JP5102181B2 - Deformed part detection method, deformed part detection device and deformed part detection program for building wall surface - Google Patents

Description

  The present invention relates to a deformed portion detecting method and a deformed portion detecting device for detecting a deformed portion of a building wall surface, and a deformed portion detecting program for operating an information processing device as the deformed portion detecting device.

  For example, buildings such as tunnels and buildings are obsolete after a long period of time from the time of construction, and it is required to properly repair the aging parts at appropriate times to extend the life of the buildings. ing. One of the manifestations of this aging is that surface layers such as the inner walls of tunnels and the outer wall of buildings are often lifted. Therefore, there is a need to prevent accidents caused by wall collapse and to reduce the labor and cost required for repairs.

  As a method for detecting such a deformed portion where the wall surface is lifted, a method for measuring the temperature distribution of the wall surface using an infrared temperature sensor is known (see, for example, Patent Document 1).

  The deformed part of the wall surface is a part where the surface layer of the wall surface is floating, and an air layer is formed on the back of the surface layer, and the air layer acts as a kind of heat insulating layer, and the outside air temperature is rising. In the middle of the outside air temperature falling, the deformed part is easier to cool before the healthy part and is more healthy. It is known that the temperature is relatively low compared to the part.

  The method of detecting the deformed portion by measuring the temperature distribution of the wall surface using an infrared temperature sensor utilizes the above-described temperature difference, and the deformed portion is typically detected by the following calculation.

  That is, using an infrared sensor, the temperature distribution of the wall surface is measured during the daytime when the outside air temperature is rising to obtain a temperature distribution image, and also when the outside air temperature at the nighttime is falling Is obtained.

  Since the temperature of the wall surface when these two temperature distribution images are obtained is generally different from each other, next, for each of the two temperature distribution images, the image value of each pixel and its temperature distribution for each pixel of each temperature distribution image Two difference images having a pixel value as a temperature difference from the average temperature obtained by calculating a difference from an average value (average temperature) around each pixel of the image are obtained.

When the pixel values of the two difference images thus determined are represented by F and G for each difference image, for each pair of pixels representing the same point on the wall surface of the two difference images, A difference D = F−G between the pixel values of the pair of pixels is calculated, and the difference is compared with a threshold T to detect a deformed portion having a temperature change different from that of the surroundings.
JP 2004-347585 A

  According to the above calculation, the temperature of the healthy part of the wall surface changes uniformly, and the deformed part of the wall temperature rises first when the outside air temperature rises and always falls first when the outside air temperature falls. It is possible to detect the deformed part under certain conditions, but on the actual wall surface, for example, a part of the wall surface becomes hot due to sunshine, or disturbance is only in the image obtained during the day. There are various disturbances such as reflection of a high-temperature object or a low-temperature object in only one image, and there is a problem that the above calculation is insufficient in accuracy. In order to compensate for this inadequate accuracy, the threshold value T is set to a level that does not miss the deformed part even if the sound part is erroneously detected as a deformed part. It is necessary to detect the deformed part including the detection and determine whether the whole part of the detected deformed part is a healthy part or a deformed part by, for example, placing a hand with a hammer. In order to make this manual judgment, for example, it is necessary to prepare an aerial work vehicle in the case of a tunnel with a large amount of traffic, and in the case of a high-rise building, it is necessary to work in a high place. Will be accompanied. For this reason, improvement in detection accuracy of the deformed portion is strongly demanded.

  Therefore, an object of the present disclosure is to improve the detection accuracy of the deformed portion.

Of the deformed portion detection method for a building wall surface of the present disclosure, the deformed portion detection method for the first building wall surface is:
The first time zone when the surface temperature of the deformed portion of the building wall surface exceeds the surface temperature of the healthy portion of the building wall surface to be detected, and the surface temperature of the deformed portion falls below the surface temperature of the sound portion. In the second time zone, two temperature distribution images having pixel values corresponding to the surface temperature of each point in the measurement area are obtained by measuring the surface temperature distribution of the same measurement area on the wall surface of the building. An image acquisition step,
For each of the two temperature distribution images, a difference for obtaining two difference images by calculating a difference between a pixel value of each pixel constituting the temperature distribution image and an average pixel value of a surrounding area of each pixel. An image generation step;
When the pixel values of the two difference images are F and G, for each pair of corresponding pixels, the difference D = F−G between the pixel values of the pair of pixels and the pixel values of the pair of pixels. An evaluation value calculating step for calculating an evaluation value E = f (D, A) represented by a function f (D, A) having a variable of the sum absolute value A = | F + G |;
And a deformed portion detecting step of detecting a deformed portion in the measurement region of the building wall surface based on the evaluation value E calculated in the evaluation value calculating step.

In addition, the second method for detecting a deformed portion on the wall surface of the building of the method for detecting a deformed portion on the wall surface of the present invention,
The first time zone during which the surface temperature of the deformed portion of the building wall surface exceeds the surface temperature of the sound portion of the building wall to be detected, and the surface temperature of the deformed portion lower than the surface temperature of the sound portion. By measuring the surface temperature distribution of the same measurement area on the wall of the building in the time period 2 and the third time period in which the surface temperatures of the healthy part and the deformed part are substantially equal, An image acquisition step of acquiring three temperature distribution images having pixel values according to the surface temperature of each point;
For each of the three temperature distribution images, a difference for obtaining three difference images by calculating a difference between a pixel value of each pixel constituting the temperature distribution image and an average pixel value of a surrounding area of each pixel. An image generation step;
Of the three difference images, the pixel values of the two difference images excluding the third difference image obtained by measurement in the third time zone are F and G, and the pixel values of the third difference image Where N is a difference F ′ = F−N between pixel values of a pair of corresponding pixels between one of the two difference images and the third difference image, and By calculating a difference G ′ = G−N between pixel values of a pair of corresponding pixels between the other difference image of the two difference images and the third difference image, two background images A background removal step for obtaining a removed image;
The difference D = F′−G ′ between the pixel values of a pair of corresponding pixels between the two background-removed images and the absolute value A = | F ′ + G ′ | of the sum of the pixel values of the pair of pixels An evaluation value calculating step for calculating an evaluation value E = f (D, A) represented by a function f (D, A) having the following variables:
And a deformed portion detecting step of detecting a deformed portion in the measurement region of the building wall surface based on the evaluation value E calculated in the evaluation value calculating step.

Moreover, the deformed part detection device of the first building wall surface among the deformed part detection devices of the building wall surface of the present disclosure is:
The first time zone when the surface temperature of the deformed portion of the building wall surface is higher than the surface temperature of the sound portion of the building wall to be detected, and the surface temperature of the deformed portion is lower than the surface temperature of the sound portion. Two temperatures having pixel values corresponding to the surface temperature of each point in the measurement region, obtained by measuring the surface temperature distribution of the same measurement region on the wall of the building during the second time period An image acquisition unit for acquiring a distribution image;
For each of the two temperature distribution images, a difference for obtaining two difference images by calculating a difference between a pixel value of each pixel constituting the temperature distribution image and an average pixel value of a surrounding area of each pixel. An image generator;
When the pixel values of the two difference images are F and G, for each pair of corresponding pixels, the difference D = F−G between the pixel values of the pair of pixels and the pixel values of the pair of pixels. An evaluation value calculation unit for calculating an evaluation value E = f (D, A) represented by a function f (D, A) having a variable of the absolute value A = | F + G |
And a deformed portion detecting unit that detects a deformed portion in the measurement region of the building wall surface based on the evaluation value E calculated by the evaluation value calculating unit.

In addition, the deformed portion detection device for the second building wall surface of the deformed portion detection device for the building wall surface of the present disclosure,
The first time zone during which the surface temperature of the deformed portion of the building wall surface exceeds the surface temperature of the sound portion of the building wall to be detected, and the surface temperature of the deformed portion lower than the surface temperature of the sound portion. Measurement obtained by measuring the surface temperature distribution in the same measurement area of the wall of the building in the time zone 2 and in the third time zone where the surface temperature of the healthy part and the deformed part are substantially equal An image acquisition unit for acquiring three temperature distribution images having pixel values according to the surface temperature of each point in the region;
For each of the three temperature distribution images, a difference image for obtaining three difference images by calculating a difference between a pixel value of each pixel constituting the temperature distribution image and an average pixel value of a surrounding area of each pixel. A generator,
Of the three difference images, the pixel values of the two difference images excluding the third difference image obtained by the measurement in the third time zone are F and G, and the pixel values of the third difference image are When N, the difference F ′ = F−N between the pixel values of a pair of corresponding pixels between one of the two difference images and the third difference image, and the two By calculating the difference G ′ = G−N between the pixel values of a pair of corresponding pixels between the other difference image and the third difference image of the difference images, the two background-removed images are obtained. The desired background removal section;
A difference D = F′−G ′ between a pair of pixels corresponding to each other between the two background-removed images, and an absolute value A = | F ′ + G ′ | An evaluation value calculation unit for calculating an evaluation value E = f (D, A) represented by a function f (D, A) having as variables;
And a deformed portion detecting unit that detects a deformed portion in the measurement region of the building wall surface based on the evaluation value E calculated by the evaluation value calculating unit.

Furthermore, the 1st deformation | transformation part detection program of the deformation | transformation part detection programs of this indication is,
The information processing apparatus is executed in an information processing apparatus that executes a program.
The first time zone when the surface temperature of the deformed portion of the building wall surface exceeds the surface temperature of the healthy portion of the building wall surface to be detected, and the surface temperature of the deformed portion falls below the surface temperature of the sound portion. Two temperature distributions having pixel values corresponding to the surface temperature of each point in the measurement region obtained by measuring the surface temperature distribution of the same measurement region on the building wall surface in the second time zone An image acquisition unit for acquiring images;
For each of the two temperature distribution images, a difference for obtaining two difference images by calculating a difference between a pixel value of each pixel constituting the temperature distribution image and an average pixel value of a surrounding area of each pixel. An image generator;
When the pixel values of the two difference images are F and G, for each pair of corresponding pixels, the difference D = F−G between the pixel values of the pair of pixels and the pixel values of the pair of pixels. An evaluation value calculation unit for calculating an evaluation value E = f (D, A) represented by a function f (D, A) having a variable of the absolute value A = | F + G |
Based on the evaluation value E calculated by the evaluation value calculation unit, the device is operated as a deformed part detecting device for a building wall having a deformed part detecting unit for detecting a deformed part in the measurement region of the building wall. It is a program.

Furthermore, the 2nd deformation | transformation part detection program of the deformation | transformation part detection programs of this indication is,
The information processing apparatus is executed in an information processing apparatus that executes a program.
The first time zone during which the surface temperature of the deformed portion of the building wall surface exceeds the surface temperature of the sound portion of the building wall to be detected, and the surface temperature of the deformed portion lower than the surface temperature of the sound portion. Measurement obtained by measuring the surface temperature distribution in the same measurement area of the wall of the building in the time zone 2 and in the third time zone where the surface temperature of the healthy part and the deformed part are substantially equal An image acquisition unit for acquiring three temperature distribution images having pixel values according to the surface temperature of each point in the region;
For each of the three temperature distribution images, a difference image generation for obtaining three difference images by calculating a difference between a pixel value of each pixel constituting the temperature distribution image and an average pixel value of a surrounding area of each pixel. And
Of the three difference images, the pixel values of the two difference images excluding the third difference image obtained by the measurement in the third time zone are F and G, and the pixel values of the third difference image Is N, the difference F ′ = F−N between the pixel values of a pair of corresponding pixels between one of the two difference images and the third difference image, and the above 2 Two background-removed images are calculated by calculating a difference G ′ = G−N between pixel values of a pair of corresponding pixels between the other difference image and the third difference image of the two difference images. A background removal unit for
A difference D = F′−G ′ between a pair of pixels corresponding to each other between the two background-removed images, and an absolute value A = | F ′ + G ′ | An evaluation value calculation unit for calculating an evaluation value E = f (D, A) represented by a function f (D, A) having as variables;
A program for operating as a deformed part detecting device for a building wall having a deformed part detecting unit for detecting a deformed part in the measurement region of the building wall based on the evaluation value E calculated by the evaluation value calculating unit It is.

  According to the present disclosure, the deformed portion can be detected with high accuracy.

  Hereinafter, an embodiment of the present case will be described.

  Here, the detection of the deformed portion of the tunnel inner wall will be described as an example.

  FIG. 1 is a diagram illustrating a schematic flow of the first embodiment of the deformed portion detection method of the present disclosure.

  The deformed portion detecting method shown in the schematic flow in FIG. 1 includes four steps of an image acquisition step S11, a difference image generating step S12, an evaluation value calculating step S13, and a deformed portion detecting step S14.

  Here, in this embodiment, the image acquisition step S11 is executed by using a temperature distribution measuring device mounted on the vehicle to measure the temperature distribution of the tunnel wall surface and obtain a temperature distribution image. Other than the image acquisition step S11 The difference image generation step S12, the evaluation value calculation step S13, and the deformed portion detection step S14 use the data processing apparatus that takes charge of the data processing after bringing back the temperature distribution image data obtained in the image acquisition step S11. Executed. Here, first, the image acquisition step S11 will be described.

  FIG. 2 is a schematic diagram showing a cross section of the inner wall of the tunnel.

  The deformed portion of the tunnel inner wall 10 is a portion where the surface layer 11 is lifted from the layer 12 deeper than the surface layer 11 and an air layer 13 is formed on the back side of the surface layer 11. In FIG. 2, the temperature distribution of the healthy part where the surface layer 11 is firmly fixed to the inner layer 12 and the air layer 13 does not exist is indicated by a solid line, and the temperature distribution of the deformed part where the air layer 13 exists is shown. Is indicated by a broken line.

  FIG. 2A is a schematic cross-sectional view showing the temperature distribution at the time when the outside air temperature is rising during the daytime. In the case of the deformed portion, the air layer 13 acts as a heat insulating layer, and the surface layer is formed by the outside air. Only 11 is warmed and the temperature rises first. On the other hand, in the case of the healthy part, since the air layer 13 does not exist, the heat transferred from the outside air to the surface layer 11 flows to the inner layer 12, and the temperature of the surface layer 11 does not rise as much as the deformed part. Is higher in temperature by Δt than the healthy part.

  FIG. 2B is a schematic cross-sectional view showing the temperature distribution at the time when the outside air temperature is decreasing at night. When the temperature of the outside air decreases, the heat of the surface layer 11 is taken away by the outside air. However, in the case of the deformed portion, the temperature of the surface layer 11 decreases first due to the presence of the air layer 13, and in the case of a healthy portion, the inner layer 12 The heat is transmitted from the surface, and the decrease in the temperature of the surface layer 11 is suppressed. For this reason, the temperature of the surface layer 11 of the deformed portion is lower than the healthy portion by Δt.

  The present case uses this phenomenon to detect the deformed portion, and this is the same as the conventional example described above.

  FIG. 3 is a diagram showing changes in the surface temperature of the healthy part and the deformed part in one day.

  The surface temperature of the deformed part is higher than that of the healthy part in the time zone A during which the outside air temperature rises during the daytime, and the surface temperature of the deformed part is higher than that of the healthy part in the time zone B during which the outside air temperature falls during the night. Is low. In addition, the intermediate time zone C1 that transitions from the time zone A to the time zone B and the intermediate time zone C2 that transits from the time zone B to the time zone A are time zones in which the surface temperatures of the healthy part and the deformed part are substantially equal. It is. Here, the temperature distribution of the tunnel inner wall is measured at any point in time zone A to obtain a temperature distribution image, and the temperature distribution of the tunnel inner wall is measured at any point in time zone B to obtain a temperature distribution image. Ask for.

  In the second embodiment to be described later, the temperature distribution of the inner wall of the tunnel is also measured in the time zone C1 or the time zone C2, and the third temperature distribution image is obtained. The second embodiment will be described later. Here, the case where two temperature distribution images in the time zones A and B are obtained will be described.

  FIG. 4 is a diagram illustrating the principle of the infrared temperature measuring device.

  In the infrared temperature measuring apparatus 20 shown in FIG. 4, a scanner mirror 21, a focus lens 22, a chopper mirror 23, an imaging lens 24, and an infrared sensor array 25 are arranged.

  The infrared sensor array 25 is a one-dimensional array of about several hundred infrared sensors. The infrared sensor array 25 receives an image of a one-dimensional image of a predetermined length (for example, 200 mm) in the horizontal direction of the inner wall of the tunnel. The original image is received by dividing it into several hundred pixels.

  The imaging lens 24 is a lens composed of a combination of a plurality of single lenses that forms a one-dimensional image of the inner wall of the tunnel on the infrared sensor array 25, and the focus lens 22 is light as the distance from the inner wall of the tunnel changes. It is a lens for focus adjustment that moves in the axial direction and focuses a one-dimensional image of the inner wall of the tunnel onto the infrared sensor array 25.

  The chopper mirror 23 will be described later.

  Furthermore, the scanner mirror 21 is a mirror that repeatedly scans the inner wall of the tunnel in the vertical direction.

  Therefore, every time the scanner mirror 21 scans the tunnel inner wall once in the vertical direction, the infrared sensor array 25 has a width of, for example, 200 mm in the horizontal direction of the tunnel inner wall, and the deflection angle of the scanner mirror 21 in the vertical direction (for example, One temperature distribution image having a length corresponding to (100 degrees) is obtained.

  The infrared temperature measuring device 20 is further provided with a plurality of reference heat sources 26a to 26f. Further, one of the reference heat sources 26a to 26f is selected, and radiant infrared rays from the selected reference heat source are received. A reference heat source lens 27 for guiding to the infrared sensor array 25 is provided. These reference heat sources 26 a to 26 f are heat sources for calibration of the infrared sensor array 25. Specifically, each of the reference heat sources 26a to 26f is precisely adjusted to different temperatures, and two reference heat sources (for example, the reference heat source 26a) sandwiching the temperature range of change of the surface temperature of the inner wall of the tunnel currently being measured. And the reference heat source 26b) are alternately selected as follows and used for calibration of the infrared sensor array 25.

  The scanner mirror 21 returns to the high speed from the top to the bottom when scanning the inner wall of the tunnel once from the bottom to the top at a predetermined speed, for example. During the return period, the chopper mirror 23 emits from the reference heat source lens 27. The infrared ray emitted from one reference heat source (in the case of FIG. 4, the reference heat source 26 b) selected by the reference heat source lens 27 is received by the infrared sensor array 25. The infrared sensor array 25 is used for calibration of several hundred infrared sensors.

  When the scanner mirror 21 scans from the top to the bottom at a high speed and then scans again from the bottom to the top, the chopper mirror 25 again connects the one-dimensional image of the inner wall of the tunnel onto the infrared sensor array 25. The scanner mirror 21 scans from the bottom to the top while changing the inclination so as to be imaged, whereby the next one temperature distribution image is obtained. Meanwhile, the reference heat source lens 27 selects the other reference heat source (for example, the reference heat source 26a), and the chopper mirror 23 is reflected again by the reference heat source mirror 27 while the scanner mirror 21 performs high-speed scanning from the top to the bottom. Infrared light emitted from one reference heat source (reference heat source 26a) selected from the reference heat source lens 27 at an angle for guiding the infrared light to the infrared sensor array 25 is received by the infrared sensor array 25 and used for calibration.

  In this way, calibration and measurement are alternately repeated, and a temperature distribution image in which the temperature distribution of the tunnel inner wall is measured with a high resolution of, for example, about 0.05 ° C. is generated. When the surface temperature of the tunnel inner wall is expected to deviate from the temperature range sandwiched between the two reference heat sources 26a and 26b, the two reference heat sources sandwiching the newly expected temperature range are selected again, and Similarly, calibration and measurement are repeated.

  FIG. 5 is a schematic diagram showing a state of temperature distribution measurement on the inner wall of the tunnel using the infrared temperature measuring apparatus having the configuration shown in FIG.

  Here, the infrared temperature measuring device 20 is installed in a vehicle 30 and the vehicle 30 travels through the tunnel at a speed of, for example, 50 km / h. During that time, the tunnel inner wall 40 is repeatedly scanned to generate a plurality of temperature distribution images. Is done. The vehicle 30 is also equipped with a distance meter 29. The distance meter 29 outputs a signal indicating that the vehicle 30 has traveled 10 mm every time the vehicle 30 has traveled 10 mm. Further, the vehicle 30 is equipped with a large-capacity storage device (not shown), and a large number of temperature distribution images (images on the data) obtained as described above are output together with an output signal from the distance meter 29. Stored in the storage device.

  In the deformed portion detection method shown in FIG. 1, the image acquisition step S <b> 11 performs the above temperature distribution image collection process at any time between time zone A and time zone B shown in FIG. 3. This step is performed at both the time and the time.

  Next, the difference image generating step S12, the evaluation value calculating step S13, and the deformed portion detecting step S14 in the deformed portion detecting method shown in FIG. 1 will be described.

  In this embodiment, the difference image generation step S12, the evaluation value calculation step S13, and the deformed portion detection step S14 are executed in the computer system by executing the deformed portion detection program in the computer system. The step to be executed. Therefore, in the following, the hardware of the computer system and the deformed portion detection program will be described in order so that the difference image generating step S12, the evaluation value calculating step S13, and the deformed portion of the deformed portion detecting method shown in FIG. The part detection step S14 will also be described.

  FIG. 6 is an external view of a computer system for performing arithmetic processing.

  The computer system 100 includes a main body 110 incorporating a CPU, a RAM memory, a hard disk, and the like, a display 120 that displays an image on a display screen 121 according to an instruction from the main body 110, and operator instructions and characters in the computer system 100. A keyboard 130 for inputting information and a mouse 140 for inputting an instruction corresponding to an icon or the like displayed at that position by designating an arbitrary position on the display screen 121 are provided.

  The main body 110 further has a CD / DVD loading slot into which a CD or a DVD (hereinafter referred to as “CD / DVD”) is loaded, and inside the loaded CD / DVD ( A CD / DVD drive 111 (see FIG. 2) that drives a drive (not shown in FIG. 1, see FIG. 2) is incorporated.

  Here, the deformed portion detection program is stored in the CD / DVD 101, and this CD / DVD 101 is loaded into the main body 110 from the CD / DVD loading slot and stored in the CD / DVD 101 by the CD / DVD drive 111. The deformed part detection program is installed in the hard disk of the computer system 100. When the deformed part detection program installed in the hard disk of the computer system is activated, the computer system 100 operates as an embodiment of the deformed part detection device.

  FIG. 7 is a hardware configuration diagram of the computer system having the appearance shown in FIG.

  Here, a central processing unit (CPU) 113, a RAM 114, a hard disk controller 115, a CD / DVD drive 111, a mouse controller 116, a keyboard controller 117, a display controller 118, and an interface 119 are provided. Are connected to each other.

  As described with reference to FIG. 6, the CD / DVD drive 111 is loaded with the CD / DVD 101 and accesses the loaded CD / DVD 101.

  The interface 119 plays a role of fetching the temperature distribution image from the storage device described above into the computer system 100.

  Also shown here are a hard disk 105 accessed by the hard disk controller 115, a mouse 140 controlled by the mouse controller 116, a keyboard 130 controlled by the keyboard controller 117, and a display 120 controlled by the display controller 118. .

  FIG. 8 is a diagram showing an outline of the deformed portion detection program.

  The deformed portion detection program 200 shown in FIG. 8 includes four program parts, an image acquisition unit 201, a difference image generation unit 202, an evaluation value calculation unit 203, and a deformed portion detection unit 204.

  Of these four program parts, the difference image generating unit 202, the evaluation value calculating unit 203, and the deformed portion detecting unit 204 are executed in the computer system 100, so that the deformed portion detecting method shown in FIG. Among these, the image acquisition unit 201 is not a step corresponding to the image acquisition step S11, but an image acquisition step, which is a program part that executes the difference image generation step S12, the evaluation value calculation step S13, and the deformed portion detection step S14. It is a program component for fetching the temperature distribution image collected as described above by execution of S11 into the computer system 100 via the interface 119 and executing processing to be described later.

  As described above, the deformed portion detection program 200 shown in FIG. 8 is stored in, for example, the CD / DVD 101, and the CD / DVD 101 is loaded into the main body 110 of the computer system 100 and is read by the CD / DVD drive 111. The deformed part detection program 200 is read and installed in the hard disk 105. When the deformed portion detection program 200 installed in the hard disk 105 is expanded on the RAM 114 and executed by the CPU 113, the computer system 100 operates as a deformed portion detecting device.

  FIG. 9 is a functional block diagram of the deformed portion detection apparatus.

  The deformed portion detection apparatus 210 includes an image acquisition unit 211, a difference image generation unit 212, an evaluation value calculation unit 213, and a deformed portion detection unit 214. The image acquisition unit 211, the difference image generation unit 212, the evaluation value calculation unit 213, and the deformed part detection unit 214 are each an image acquisition unit that is a program component constituting the deformed part detection program 200 illustrated in FIG. 6. 201, the difference image generation unit 202, the evaluation value calculation unit 203, and the deformed part detection unit 204 are functions built in the computer system 100 by being executed in the computer system 100. By explaining the operation of each part 211 to 214 of the deformed part detection device 210 of FIG. 9, the parts 201 to 204 as the respective program parts constituting the deformed part detection program 200 shown in FIG. To do. Further, by explaining each of the other parts 212 to 214 excluding the image acquiring unit 211 of the deformed part detecting device 210 of FIG. 9, each of the deformed part detecting methods shown in FIG. 1 except for the image acquiring step S11. Steps S12 to S14 will also be described together.

  The image acquisition unit 211 of the deformed part detection device 210 shown in FIG. 9 stores the temperature distribution image collected as described above and stored in a storage device (not shown) in the computer system 100 via the interface 119. And various geometric corrections.

  As the geometric correction, correction of a deviation in the traveling direction of an image caused by a change in traveling speed of the vehicle 30 shown in FIG. 5, a change in a traveling position of the vehicle 30 in the width direction of the road, that is, the vehicle 30. And correction of enlargement / reduction of an image due to a change in the distance between the temperature distribution measuring device 20 mounted on the tunnel and the inner wall of the tunnel. The correction of the deviation in the traveling direction of the image is corrected based on a signal output every 10 mm traveling from the distance meter 29, and for enlargement / reduction, pattern matching of a plurality of images obtained from the same area of the tunnel inner wall, etc. It is corrected by. By such correction, two temperature distribution images obtained by temperature distribution measurement in each of the two time zones A and B shown in FIG. 3, and tunneling is performed for each pixel corresponding to the two temperature distribution images. Two temperature distribution images representing the temperature at the same point on the inner wall are generated.

  Here, the geometric correction itself may apply a conventional method, and further detailed description thereof is omitted here. Although the above description has been made on the assumption that the image acquisition unit 211 performs the above correction, the correction is made before the temperature distribution image is taken into the deformed portion detection device 210 (computer system 100 shown in FIGS. 6 and 7). In other words, the image acquisition unit 211 may capture the corrected temperature distribution image as a role of the image acquisition step S11 in the deformed portion detection method shown in FIG.

  In the difference image generation unit 212 of the deformed portion detection apparatus 210 shown in FIG. 9, the temperature distribution image for each of the two temperature distribution images obtained by the image acquisition unit 211 after the geometric correction is performed. The difference between the pixel value of each pixel constituting the pixel and the average pixel value of the surrounding area of each pixel is calculated, and thereby two difference images are obtained. The area of the peripheral area for obtaining the average pixel value is the average value for the entire temperature distribution image when the surface temperature of the tunnel inner wall corresponding to the entire temperature distribution image is relatively stable. Or, when the stability is relatively low, it may be an average value of a partial area around each pixel, or statistically examine the distribution of pixel values around each pixel, and The width of the surrounding area for obtaining the average value may be adaptively changed according to the distribution. In any case, the calculation is performed so that the pixel value of each pixel (the surface temperature of one point on the inner wall of the tunnel corresponding to the pixel) is the pixel value. Here, the pixel values of the two difference images obtained in this way are represented by F and G, representing each difference image.

  In the present embodiment, the pixel value of the difference image obtained from the temperature distribution image obtained by the measurement in the time zone A of FIG. 3 is F, and the pixel of the difference image obtained from the temperature distribution image obtained by the measurement in the time zone B Let G be the value. Each of the pixel values F and G represents a difference from the average surface temperature of the surroundings, and therefore can take either plus / minus values. As can be seen from FIG. 3, the pixel value F typically takes a positive value and the pixel value G takes a negative value for the deformed portion.

When the difference image generation unit 212 obtains two difference images as described above, the evaluation value calculation unit 213 next obtains an evaluation value for discriminating between a healthy part and a deformed part. In the difference image generation unit 212, for each pair of pixels corresponding to each other in the two difference images, the difference D = F−G between the pixel values of the pair of pixels and the absolute value of the sum of the pixel values of the pair of pixels. A = | F + G | is calculated, and for each pixel, an evaluation value E = f (D, A) represented by a function f (D, A) having the difference D and the absolute value A of the variables as variables is obtained. Calculated. Typically, as this evaluation value E, E = C · (D−A) (where C is a proportional constant, and 0.5 is adopted in the present embodiment. E = D is adopted as E, and the constant C = 0.5 is for adjusting the scale with the conventional evaluation value E = D, for comparison with the conventional method. Since it is not always necessary to adjust the scale for the purpose of detecting the deformed portion, C may be a constant having a value other than 0.5, and in the conventional method, the difference D is used as the evaluation value E. However, the evaluation value E is not limited to E = C · (D−A), for example. Evaluation value weighted between the difference D and the absolute value A of the sum E = C · (d1 · D−d2 · A)
However, d1 + d2 = 1.0
May be adopted, or a more complicated arithmetic expression may be adopted.

  Here, E = 0.5 · (DA) = (DA) / 2 is adopted as the evaluation value E, and the description will be continued.

  In the deformed portion detecting unit 214 of the deformed portion detecting apparatus 210 shown in FIG. 9, the deformed portion in the measurement region of the tunnel inner wall is detected based on the evaluation value E calculated by the evaluation value calculating unit 213. Specifically, the evaluation value E for each pixel calculated by the evaluation value calculation unit 213 is compared with the threshold T, and the pixel for which the evaluation value E exceeding the threshold T is calculated is a pixel corresponding to the deformed portion. It is determined. The threshold T is set to an appropriate value empirically or statistically according to the material of the tunnel inner wall.

  Here, a combination of pixel values F and G of two difference images will be described as an example of calculation for detecting a deformed portion.

  Here, the pixel value of a pixel having a higher temperature than the surrounding area is +10, the average pixel value representing the background area, that is, the healthy area is 0, and the pixel value of a low temperature area, that is, a pixel having a temperature lower than the surrounding area is −10. To do.

At this time, possible combinations of the pixel values F and G are:
F: High temperature / G: High temperature F: High temperature / G: Background F: High temperature / G: Low temperature F: Background / G: High temperature F: Background / G: Background F: Background / G: Low temperature F: Low temperature / G: High temperature F : Low temperature / G: Background F: Low temperature / G: Low temperature

Nine ways.

  When the above calculation is performed for these nine combinations, the results are shown in Table 1.

  When E = (D−A) / 2 is adopted as the evaluation value E and 5 is adopted as the threshold value T, the pixel c whose pixel value F is higher than the background and whose pixel value G is lower than the display. It can be seen that only is extracted.

  In the conventional method, E = D is adopted as the evaluation value E. In this case, the pixels k and f are also extracted as the value of the pixel c. These pixels b and f are considered to correspond to a case where a high-temperature object or a low-temperature object is reflected in one of two temperature distribution pixels from which the pixel values F and G are obtained. Further, the pixel b may have a case where a part of the image obtained by the measurement in the time zone A shown in FIG.

  Next, the effect when the temperature changes in the high temperature part and the low temperature part are different from each other will be described. When the pixel value in the high temperature part is +12 and the pixel value in the low temperature part is -8, the following Table 2 is obtained.

  Looking at the column of D = F−G in Table 2, a value of 12 is obtained for pixel b, 20 for pixel C, and 8 for pixel f. In the case of Table 1, the pixels b, c, and f are 10, 20, and 10, respectively. If the evaluation value E = D is used to obtain a detection result similar to that of the present embodiment, the threshold value T = 11. If the threshold value T = 11, if the temperature change in the high temperature part and the temperature change in the low temperature part are different from +12 and −8 as shown in Table 2, not only the pixel c but also the pixel b is detected. At that time, the determination result varies depending on a slight condition change at that time.

  On the other hand, in the case of this embodiment, even if the temperature change of a high temperature part and a low temperature part differs, a deformed part can be detected stably with high accuracy.

  FIG. 10 is a diagram illustrating an example of the deformed portion detection result.

  FIG. 10A shows a region detected as a deformed portion when the difference D is adopted as the evaluation value E in the conventional method, and FIG. 10B shows E = (DA) as the evaluation value E. An area detected as a deformed portion when / 2 is adopted is shown. Since the scale of the evaluation value E is adjusted by dividing (D−A) by 2 (multiplying by 0.5), the same value is used as the threshold value in FIGS. 10 (A) and 10 (B). Adopted.

  Comparing FIGS. 10A and 10B, the conventional method includes many erroneous detections, and a wide area is detected as the deformed portion. In this embodiment, the deformed portion is increased by reducing the false detection. It can be seen that it is detected accurately.

  Next, a second embodiment of the present case will be described.

  FIG. 11 is a flowchart showing an outline of the deformed portion detection method of the second embodiment of the present case.

  The deformed portion detection method according to the second embodiment includes image acquisition step S21, difference image generation step S22, background removal step S23, evaluation value calculation step S24, and deformed portion detection step S25. .

  The description will focus on the differences from the first embodiment described above.

  In the image acquisition step S21 of the deformed portion detection method of the second embodiment shown in FIG. 11, the temperature distribution of the tunnel inner wall is measured in both time zones A and B shown in FIG. 3 to obtain two temperature distribution images. In addition, the temperature distribution measurement of the inner wall of the tunnel is also performed in the time zone C1 or the time zone C2 in which the temperatures of the healthy portion and the deformed portion are the same, and a third temperature distribution image is obtained. Other points are the same as in the case of the first embodiment described above.

  Other steps S22 to S25 of the deformed portion detecting method of the second embodiment shown in FIG. 11 will be described later together with the deformed portion detecting program and deformed portion detecting device of the second embodiment.

  FIG. 12 is a diagram showing an outline of the deformed portion detection program of the second embodiment.

  The deformed portion detection program 300 of the second embodiment shown in FIG. 12 includes five parts, an image acquisition unit 301, a difference image generation unit 302, a background removal unit 303, an evaluation value calculation unit 304, and a deformed portion detection unit 305. It consists of program parts. Among these five program parts, the difference image generation unit 302, the background removal unit 303, the evaluation value calculation unit 304, and the deformed part detection unit 305 are executed in the computer system 100, thereby changing the change shown in FIG. The image acquisition unit 301 executes the difference image generation step S22, the background removal step S23, the evaluation value calculation step S24, and the deformed portion detection step S25 in the shape detection method. Is a program component for fetching and processing the temperature distribution images of the three time zones collected by the execution of the image acquisition step S21 into the computer system 100 via the interface 119.

  The deformed portion detection program 300 shown in FIG. 12 is also stored in the CD / DVD 101, for example, like the deformed portion detection program 200 of the first embodiment shown in FIG. 8, and the CD / DVD 101 is stored in the computer. The deformed part detection program 200 is loaded into the main body 110 of the system 100 and read by the CD / DVD drive 111 and installed in the hard disk 105. When the deformed portion detection program 300 installed in the hard disk 105 is expanded on the RAM 114 and executed by the CPU 113, the computer system 100 operates as the deformed portion detecting device of the second embodiment.

  FIG. 13 is a block diagram showing the configuration of the deformed portion detection apparatus according to the second embodiment of the present invention.

  The deformed portion detection apparatus 310 includes an image acquisition unit 311, a difference image generation unit 312, a background removal unit 313, an evaluation value calculation unit 314, and a deformed portion detection unit 315. The image acquisition unit 311, the difference image generation unit 312, the background removal unit 313, the evaluation value calculation unit 314, and the deformed part detection unit 315 are program components constituting the deformed part detection program 300 shown in FIG. An image acquisition unit 301, a difference image generation unit 302, a background removal unit 303, an evaluation value calculation unit 304, and a deformed part detection unit 305 are executed in the computer system 100 to be constructed in the computer system 100. It is a function. Accordingly, as in the case of the first embodiment, the operation of each unit 311 to 315 of the deformed portion detection apparatus 310 in FIG. 13 will be described below as each program component constituting the deformed portion detection program 300 shown in FIG. These parts 301 to 305 will be described together. Further, by explaining each of the other parts 312 to 315 except for the image acquiring unit 311 of the deformed part detecting device 310 of FIG. 13, each step S22 excluding the image acquiring step S21 in the deformed part detecting method shown in FIG. ˜S25 will also be described together. Here, differences from the first embodiment will be described.

  An image acquisition unit 311 of the deformed part detection device 310 illustrated in FIG. 13 is the same as the image acquisition unit 211 of the deformed part detection device 210 (see FIG. 9) of the first embodiment. However, in the image acquisition unit 311 of the deformed part detection device 310 shown in FIG. 13, in addition to the temperature distribution image obtained by the measurement in the time zones A and B, it is obtained by the measurement in the time zone C1 or the time zone C2. The third temperature distribution image is also taken into the computer system 100 and subjected to geometric correction.

  The operation of the difference image generation unit 312 of the deformed portion detection apparatus 300 in FIG. 13 is also the same as that of the difference image generation unit 212 of the deformed portion detection apparatus 210 (see FIG. 9) of the first embodiment. However, the difference image generation unit 312 of the deformed portion detection apparatus 310 generates a third difference image in the same manner for the third temperature distribution image obtained by the measurement in the time zone C1 or the time zone C2. The Here, the pixel value of the difference image obtained from the temperature distribution image obtained by the measurement in the time zone A is represented by F, and the pixel value is obtained by the difference image obtained from the temperature distribution image obtained by the measurement in the time zone B. As a representative, the pixel value of the third difference image obtained from the third temperature distribution image obtained by the measurement in G, time zone C1 or time zone C2 is represented by N.

  Next, in the background removal unit 313 of the deformed part detection device 310 illustrated in FIG. 13, the first difference image derived from the measurement in the time zone A and the third difference derived from the measurement in the time zone C1 or the time zone C2. For the image, the difference F ′ = F−N between the pixel values F and N is obtained for each pixel corresponding to each other, and similarly, the second difference image derived from the measurement in the time zone B and the time zone C1 or For the third difference image derived from the measurement in the time zone C2, the difference G ′ = G−N between the pixel values G and N is obtained for each corresponding pixel.

  The temperature distribution image obtained by measuring the infrared rays emitted from the tunnel inner wall does not accurately represent the surface temperature distribution of the tunnel inner wall, and the pixel value of the temperature distribution image is also affected by the pattern and color of the tunnel inner wall. Receive. As described above, when the third difference image derived from the measurement in the time zone in which the temperature of the healthy part and the deformed part is substantially the same, the third difference image is the healthy part or the deformed part. The image does not matter. So, in this 2nd Embodiment, the 3rd difference image derived from the measurement of the time zone in which the temperature of a healthy part and a deformed part is substantially the same is calculated | required, and two differences derived from the measurement of the time zones A and B By subtracting from the image, the influence of the pattern and color of the tunnel inner wall is removed. By doing so, the deformed portion can be detected with higher accuracy.

  The operation of the evaluation value calculation unit 314 and the deformation part detection unit 315 constituting the deformation part detection device 310 shown in FIG. 13 is the same as the evaluation value calculation part 213 and the deformation part that constitute the deformation part detection device 210 shown in FIG. The operation of the part detection unit 214 is the same. However, the evaluation value calculation unit 213 of the deformed part detection device 210 shown in FIG. 9 obtains the evaluation value E using the pixel values F and G as they are, whereas the evaluation of the deformed part detection device 310 shown in FIG. In the value calculation unit 314, instead of using the pixel values F and G as they are, the evaluation value E is obtained by using the pixel values F ′ and G ′ obtained by the background removal unit 313.

  FIG. 14 is a diagram illustrating an example of the deformed portion detection result.

  FIG. 14A shows the detection result in the first embodiment described above, and represents the same detection result as in FIG. FIG. 14B shows a detection result when the second embodiment is adopted. In particular, a region surrounded by a broken line disappears, false detection is further reduced, and only the deformed portion is detected with higher accuracy. .

  In this case, the deformed portion is detected by setting only one threshold T. However, the threshold T is set in a plurality of steps such as two steps, and the region that is the deformed portion and the deformed portion are suspected. It may be determined and displayed or notified by distinguishing the areas to be displayed.

  Further, here, the detection of the deformed portion of the tunnel inner wall has been described, but the present case is also applicable to the detection of the deformed portion of the outer wall of a building such as a building or a house. Further, when the temperature distribution measuring device is fixed and the environment is capable of measuring the temperature distribution in a plurality of time zones, the geometric correction of the image is unnecessary.

  Hereinafter, various forms of the present case will be added.

(Appendix 1)
The first time zone in which the surface temperature of the deformed portion of the building wall surface exceeds the surface temperature of the sound portion of the building wall to be detected, and the surface temperature of the deformed portion than the surface temperature of the sound portion In the second time zone where the value falls below, the surface temperature distribution of the same measurement area on the wall of the building is measured to obtain two pixel values corresponding to the surface temperature of each point in the measurement area. An image acquisition step for acquiring a temperature distribution image;
For each of the two temperature distribution images, two difference images are obtained by calculating the difference between the pixel value of each pixel constituting the temperature distribution image and the average pixel value of the surrounding area of each pixel. A difference image generation step;
When the pixel values of the two difference images are F and G, for each pair of corresponding pixels, the difference D = F−G between the pixel values of the pair of pixels and the pixel value of the pair of pixels An evaluation value calculating step of calculating an evaluation value E = f (D, A) represented by a function f (D, A) having a variable of an absolute value A = | F + G |
A deformed portion detecting step of detecting a deformed portion in the measurement area of the building wall surface based on the evaluation value E calculated in the evaluation value calculating step. -Like part detection method.

(Appendix 2)
In the first time zone in which the surface temperature of the deformed portion of the building wall surface exceeds the surface temperature of the sound portion of the building wall to be detected, the surface temperature of the deformed portion is higher than the surface temperature of the sound portion. Measuring the surface temperature distribution in the same measurement area of the wall of the building in the second time zone that falls below and the third time zone in which the surface temperatures of the healthy part and the deformed part are substantially equal. An image acquisition step of acquiring three temperature distribution images having pixel values according to the surface temperature of each point in the measurement region;
For each of the three temperature distribution images, three difference images are obtained by calculating the difference between the pixel value of each pixel constituting the temperature distribution image and the average pixel value of the surrounding area of each pixel. A difference image generation step;
Of the three difference images, the pixel values of the two difference images excluding the third difference image obtained by measurement in the third time zone are F and G, and the pixel values of the third difference image Where N is a difference F ′ = F−N between pixel values of a pair of corresponding pixels between one of the two difference images and the third difference image, and By calculating a difference G ′ = G−N between pixel values of a pair of corresponding pixels between the other difference image of the two difference images and the third difference image, two background images A background removal step for obtaining a removed image;
The difference D = F′−G ′ between the pixel values of a pair of corresponding pixels between the two background-removed images and the absolute value A = | F ′ + G ′ | of the sum of the pixel values of the pair of pixels An evaluation value calculating step for calculating an evaluation value E = f (D, A) represented by a function f (D, A) having the following variables:
A deformed portion detecting step of detecting a deformed portion in the measurement area of the building wall surface based on the evaluation value E calculated in the evaluation value calculating step. -Like part detection method.

(Appendix 3)
In the evaluation value calculating step, when the proportionality constant is C, as the evaluation value E,
E = C · (DA)
Is the step of calculating
Supplementary note 1 or 2, wherein the deformed portion detecting step is a step of detecting the deformed portion by comparing the evaluation value E calculated in the evaluation value calculating step with a predetermined threshold T. Method for detecting deformed part of building wall.

(Appendix 4)
The first time zone in which the surface temperature of the deformed portion of the building wall surface exceeds the surface temperature of the sound portion of the building wall to be detected, and the surface temperature of the deformed portion than the surface temperature of the sound portion The pixel value corresponding to the surface temperature of each point in the measurement area obtained by measuring the surface temperature distribution of the same measurement area on the wall surface of the building during the second time zone when An image acquisition unit for acquiring two temperature distribution images,
For each of the two temperature distribution images, two difference images are obtained by calculating the difference between the pixel value of each pixel constituting the temperature distribution image and the average pixel value of the surrounding area of each pixel. A difference image generation unit;
When the pixel values of the two difference images are F and G, for each pair of corresponding pixels, the difference D = F−G between the pixel values of the pair of pixels and the pixel value of the pair of pixels An evaluation value calculation unit for calculating an evaluation value E = f (D, A) represented by a function f (D, A) having the absolute value A = | F + G |
A deformed portion detecting unit that detects a deformed portion in the measurement region of the building wall surface based on the evaluation value E calculated by the evaluation value calculating unit. Deformed part detection device.

(Appendix 5)
In the first time zone in which the surface temperature of the deformed portion of the building wall surface exceeds the surface temperature of the sound portion of the building wall to be detected, the surface temperature of the deformed portion is higher than the surface temperature of the sound portion. Measuring the surface temperature distribution in the same measurement area of the wall of the building in the second time zone that falls below and the third time zone in which the surface temperatures of the healthy part and the deformed part are substantially equal. An image acquisition unit for acquiring three temperature distribution images having pixel values according to the surface temperature of each point in the measurement region, obtained by
For each of the three temperature distribution images, a difference for obtaining three difference images by calculating a difference between a pixel value of each pixel constituting the temperature distribution image and an average pixel value of a surrounding area of each pixel. An image generator;
Of the three difference images, the pixel values of the two difference images excluding the third difference image obtained by measurement in the third time zone are F and G, and the pixel values of the third difference image Where N is a difference F ′ = F−N between pixel values of a pair of corresponding pixels between one of the two difference images and the third difference image, and By calculating a difference G ′ = G−N between pixel values of a pair of corresponding pixels between the other difference image of the two difference images and the third difference image, two background images A background removal unit for obtaining a removed image;
The difference D = F′−G ′ between the pixel values of a pair of corresponding pixels between the two background-removed images and the absolute value A = | F ′ + G ′ | of the sum of the pixel values of the pair of pixels An evaluation value calculation unit for calculating an evaluation value E = f (D, A) represented by a function f (D, A) having variables as variables,
A deformed portion detecting unit that detects a deformed portion in the measurement region of the building wall surface based on the evaluation value E calculated by the evaluation value calculating unit. Deformed part detection device.

(Appendix 6)
When the evaluation value calculation unit sets the proportionality constant to C, the evaluation value E
E = C · (DA)
Is calculated,
The appendix 4 or 5, wherein the deformed portion detecting unit detects the deformed portion by comparing the evaluation value E calculated by the evaluation value calculating unit with a predetermined threshold T. Deformation detection device for building walls.

(Appendix 7)
The information processing apparatus is executed in an information processing apparatus that executes a program.
The first time zone in which the surface temperature of the deformed portion of the building wall surface exceeds the surface temperature of the sound portion of the building wall to be detected, and the surface temperature of the deformed portion than the surface temperature of the sound portion The pixel value corresponding to the surface temperature of each point in the measurement area obtained by measuring the surface temperature distribution of the same measurement area on the wall surface of the building during the second time zone when An image acquisition unit for acquiring two temperature distribution images,
For each of the two temperature distribution images, two difference images are obtained by calculating the difference between the pixel value of each pixel constituting the temperature distribution image and the average pixel value of the surrounding area of each pixel. A difference image generation unit;
When the pixel values of the two difference images are F and G, for each pair of corresponding pixels, the difference D = F−G between the pixel values of the pair of pixels and the pixel value of the pair of pixels An evaluation value calculation unit for calculating an evaluation value E = f (D, A) represented by a function f (D, A) having the absolute value A = | F + G |
As a deformed part detecting device for a building wall, comprising a deformed part detecting unit for detecting a deformed part in the measurement area of the building wall based on the evaluation value E calculated by the evaluation value calculating part. A program for detecting a deformed portion of a building wall characterized by being operated.

(Appendix 8)
The information processing apparatus is executed in an information processing apparatus that executes a program.
In the first time zone in which the surface temperature of the deformed portion of the building wall surface exceeds the surface temperature of the sound portion of the building wall to be detected, the surface temperature of the deformed portion is higher than the surface temperature of the sound portion. Measuring the surface temperature distribution in the same measurement area of the wall of the building in the second time zone that falls below and the third time zone in which the surface temperatures of the healthy part and the deformed part are substantially equal. An image acquisition unit for acquiring three temperature distribution images having pixel values according to the surface temperature of each point in the measurement region, obtained by
For each of the three temperature distribution images, a difference for obtaining three difference images by calculating a difference between a pixel value of each pixel constituting the temperature distribution image and an average pixel value of a surrounding area of each pixel. An image generator;
Of the three difference images, the pixel values of the two difference images excluding the third difference image obtained by measurement in the third time zone are F and G, and the pixel values of the third difference image Where N is a difference F ′ = F−N between pixel values of a pair of corresponding pixels between one of the two difference images and the third difference image, and By calculating a difference G ′ = G−N between pixel values of a pair of corresponding pixels between the other difference image of the two difference images and the third difference image, two background images A background removal unit for obtaining a removed image;
The difference D = F′−G ′ between the pixel values of a pair of corresponding pixels between the two background-removed images and the absolute value A = | F ′ + G ′ | of the sum of the pixel values of the pair of pixels An evaluation value calculation unit for calculating an evaluation value E = f (D, A) represented by a function f (D, A) having variables
As a deformed part detecting device for a building wall, comprising a deformed part detecting unit for detecting a deformed part in the measurement area of the building wall based on the evaluation value E calculated by the evaluation value calculating part. A program for detecting a deformed portion of a building wall characterized by being operated.

(Appendix 9)
When the evaluation value calculation unit sets the proportionality constant to C, the evaluation value E
E = C · (DA)
Is calculated,
The added part 7 or 8, wherein the deformed part detecting part detects the deformed part by comparing the evaluation value E calculated by the evaluation value calculating part with a predetermined threshold T. Deformation detection program for building walls.

It is a figure which shows the schematic flow of 1st Embodiment of the deformation | transformation part detection method of this indication. It is a schematic diagram which shows the cross section of a tunnel inner wall. It is the figure which showed the change of the surface temperature of the healthy part and deformed part in one day. It is principle explanatory drawing of an infrared temperature measuring apparatus. It is the schematic diagram which showed the mode of the temperature distribution measurement of a tunnel inner wall using an infrared temperature measuring apparatus. It is an external view of a computer system for performing arithmetic processing. It is a hardware block diagram of the computer system which has the external appearance shown in FIG. It is a figure which shows the outline | summary of a deformation | transformation part detection program. It is a functional block diagram of a deformation | transformation part detection apparatus. It is a figure which shows an example of a deformation | transformation part detection result. It is a flowchart which shows the outline | summary of the deformation | transformation part detection method of 2nd Embodiment. It is a figure which shows the outline | summary of the deformation | transformation part detection program of 2nd Embodiment. It is a block diagram of the configuration of the deformed portion detection device of the second embodiment. It is a figure which shows an example of a deformation | transformation part detection result.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10,40 Tunnel inner wall 11 Surface layer 12 Back layer 13 Air layer 20 Infrared temperature measuring device 21 Scanner mirror 22 Focus lens 23 Chopper mirror 24 Imaging lens 25 Infrared sensor array 26 Reference heat source 27 Reference heat source lens
29 Distance meter 30 Vehicle 100 Computer system 101 CD / DVD
105 Hard Disk 109 Bus 110 Main Body 111 CD / DVD Drive 112 MO Drive 113 CPU
114 RAM
115 Hard Disk Controller 116 Mouse Controller 117 Keyboard Controller 118 Display Controller 119 Interface 120 Display 121 Display Screen 130 Keyboard 140 Mouse 200,300 Deformed Section Detection Program 201, 211, 301, 311 Image Acquisition Unit 202, 212, 302, 312 Difference Image Generation unit 203, 213, 304, 314 Evaluation value calculation unit 204, 214, 305, 315 Deformation part detection unit 210, 310 Deformation part detection device 303, 313 Background removal unit

Claims (6)

The first time zone in which the surface temperature of the deformed portion of the building wall surface exceeds the surface temperature of the sound portion of the building wall to be detected, and the surface temperature of the deformed portion than the surface temperature of the sound portion In the second time zone where the value falls below, the surface temperature distribution of the same measurement area on the wall of the building is measured to obtain two pixel values corresponding to the surface temperature of each point in the measurement area. An image acquisition step for acquiring a temperature distribution image;
For each of the two temperature distribution images, two difference images are obtained by calculating the difference between the pixel value of each pixel constituting the temperature distribution image and the average pixel value of the surrounding area of each pixel. A difference image generation step;
When the pixel values of the two difference images are F and G, for each pair of corresponding pixels, the difference D = F−G between the pixel values of the pair of pixels and the pixel value of the pair of pixels An evaluation value calculating step of calculating an evaluation value E = f (D, A) represented by a function f (D, A) having a variable of an absolute value A = | F + G |
A deformed portion detecting step of detecting a deformed portion in the measurement area of the building wall surface based on the evaluation value E calculated in the evaluation value calculating step. -Like part detection method. In the first time zone in which the surface temperature of the deformed portion of the building wall surface exceeds the surface temperature of the sound portion of the building wall to be detected, the surface temperature of the deformed portion is higher than the surface temperature of the sound portion. Measuring the surface temperature distribution in the same measurement area of the wall of the building in the second time zone that falls below and the third time zone in which the surface temperatures of the healthy part and the deformed part are substantially equal. An image acquisition step of acquiring three temperature distribution images having pixel values according to the surface temperature of each point in the measurement region;
For each of the three temperature distribution images, three difference images are obtained by calculating the difference between the pixel value of each pixel constituting the temperature distribution image and the average pixel value of the surrounding area of each pixel. A difference image generation step;
Of the three difference images, the pixel values of the two difference images excluding the third difference image obtained by measurement in the third time zone are F and G, and the pixel values of the third difference image Where N is a difference F ′ = F−N between pixel values of a pair of corresponding pixels between one of the two difference images and the third difference image, and By calculating a difference G ′ = G−N between pixel values of a pair of corresponding pixels between the other difference image of the two difference images and the third difference image, two background images A background removal step for obtaining a removed image;
The difference D = F′−G ′ between the pixel values of a pair of corresponding pixels between the two background-removed images and the absolute value A = | F ′ + G ′ | of the sum of the pixel values of the pair of pixels An evaluation value calculating step for calculating an evaluation value E = f (D, A) represented by a function f (D, A) having the following variables:
A deformed portion detecting step of detecting a deformed portion in the measurement area of the building wall surface based on the evaluation value E calculated in the evaluation value calculating step. -Like part detection method. The first time zone in which the surface temperature of the deformed portion of the building wall surface exceeds the surface temperature of the sound portion of the building wall to be detected, and the surface temperature of the deformed portion than the surface temperature of the sound portion The pixel value corresponding to the surface temperature of each point in the measurement area obtained by measuring the surface temperature distribution of the same measurement area on the wall surface of the building during the second time zone when An image acquisition unit for acquiring two temperature distribution images,
For each of the two temperature distribution images, two difference images are obtained by calculating the difference between the pixel value of each pixel constituting the temperature distribution image and the average pixel value of the surrounding area of each pixel. A difference image generation unit;
When the pixel values of the two difference images are F and G, for each pair of corresponding pixels, the difference D = F−G between the pixel values of the pair of pixels and the pixel value of the pair of pixels An evaluation value calculation unit for calculating an evaluation value E = f (D, A) represented by a function f (D, A) having the absolute value A = | F + G |
A deformed portion detecting unit that detects a deformed portion in the measurement region of the building wall surface based on the evaluation value E calculated by the evaluation value calculating unit. Deformed part detection device. In the first time zone in which the surface temperature of the deformed portion of the building wall surface exceeds the surface temperature of the sound portion of the building wall to be detected, the surface temperature of the deformed portion is higher than the surface temperature of the sound portion. Measuring the surface temperature distribution in the same measurement area of the wall of the building in the second time zone that falls below and the third time zone in which the surface temperatures of the healthy part and the deformed part are substantially equal. An image acquisition unit for acquiring three temperature distribution images having pixel values according to the surface temperature of each point in the measurement region, obtained by
For each of the three temperature distribution images, a difference for obtaining three difference images by calculating a difference between a pixel value of each pixel constituting the temperature distribution image and an average pixel value of a surrounding area of each pixel. An image generator;
Of the three difference images, the pixel values of the two difference images excluding the third difference image obtained by measurement in the third time zone are F and G, and the pixel values of the third difference image Where N is a difference F ′ = F−N between pixel values of a pair of corresponding pixels between one of the two difference images and the third difference image, and By calculating a difference G ′ = G−N between pixel values of a pair of corresponding pixels between the other difference image of the two difference images and the third difference image, two background images A background removal unit for obtaining a removed image;
The difference D = F′−G ′ between the pixel values of a pair of corresponding pixels between the two background-removed images and the absolute value A = | F ′ + G ′ | of the sum of the pixel values of the pair of pixels An evaluation value calculation unit for calculating an evaluation value E = f (D, A) represented by a function f (D, A) having variables as variables,
A deformed portion detecting unit that detects a deformed portion in the measurement region of the building wall surface based on the evaluation value E calculated by the evaluation value calculating unit. Deformed part detection device. The information processing apparatus is executed in an information processing apparatus that executes a program.
The first time zone in which the surface temperature of the deformed portion of the building wall surface exceeds the surface temperature of the sound portion of the building wall to be detected, and the surface temperature of the deformed portion than the surface temperature of the sound portion The pixel value corresponding to the surface temperature of each point in the measurement area obtained by measuring the surface temperature distribution of the same measurement area on the wall surface of the building during the second time zone when An image acquisition unit for acquiring two temperature distribution images,
For each of the two temperature distribution images, two difference images are obtained by calculating the difference between the pixel value of each pixel constituting the temperature distribution image and the average pixel value of the surrounding area of each pixel. A difference image generation unit;
When the pixel values of the two difference images are F and G, for each pair of corresponding pixels, the difference D = F−G between the pixel values of the pair of pixels and the pixel value of the pair of pixels An evaluation value calculation unit for calculating an evaluation value E = f (D, A) represented by a function f (D, A) having the absolute value A = | F + G |
As a deformed part detecting device for a building wall, comprising a deformed part detecting unit for detecting a deformed part in the measurement area of the building wall based on the evaluation value E calculated by the evaluation value calculating part. A program for detecting a deformed portion of a building wall characterized by being operated. The information processing apparatus is executed in an information processing apparatus that executes a program.
In the first time zone in which the surface temperature of the deformed portion of the building wall surface exceeds the surface temperature of the sound portion of the building wall to be detected, the surface temperature of the deformed portion is higher than the surface temperature of the sound portion. Measuring the surface temperature distribution in the same measurement area of the wall of the building in the second time zone that falls below and the third time zone in which the surface temperatures of the healthy part and the deformed part are substantially equal. An image acquisition unit for acquiring three temperature distribution images having pixel values according to the surface temperature of each point in the measurement region, obtained by
For each of the three temperature distribution images, a difference for obtaining three difference images by calculating a difference between a pixel value of each pixel constituting the temperature distribution image and an average pixel value of a surrounding area of each pixel. An image generator;
Of the three difference images, the pixel values of the two difference images excluding the third difference image obtained by measurement in the third time zone are F and G, and the pixel values of the third difference image Where N is a difference F ′ = F−N between pixel values of a pair of corresponding pixels between one of the two difference images and the third difference image, and By calculating a difference G ′ = G−N between pixel values of a pair of corresponding pixels between the other difference image of the two difference images and the third difference image, two background images A background removal unit for obtaining a removed image;
The difference D = F′−G ′ between the pixel values of a pair of corresponding pixels between the two background-removed images and the absolute value A = | F ′ + G ′ | of the sum of the pixel values of the pair of pixels An evaluation value calculation unit for calculating an evaluation value E = f (D, A) represented by a function f (D, A) having variables as variables,
As a deformed part detecting device for a building wall, comprising a deformed part detecting unit for detecting a deformed part in the measurement area of the building wall based on the evaluation value E calculated by the evaluation value calculating part. A program for detecting a deformed portion of a building wall characterized by being operated.

Images