Infrared camera, image processing method, and image processing program

Abstract

To reduce noise in a captured image.SOLUTION: An infrared camera 10 includes an execution unit 32 that executes image processing on a captured image. The execution unit 32 includes an acquisition unit 322 that acquires a noise component extracted from the captured image and the captured image obtained by capturing the portion covered by a black body, and a removal unit 324 that removes the noise component from the entire captured image.SELECTED DRAWING: Figure 5

Description

Technical Field

[0001] The present invention relates to an infrared camera, an image processing method, and an image processing program.

Background Art

[0002] Patent Document 1 discloses a method for investigating a structure using an infrared camera, which photographs the surface of a structure with an infrared camera and investigates the damage state inside the structure based on the temperature distribution on the photographed surface of the structure. In this method for investigating a structure, a plurality of damage patterns with different depths and shapes of damage parts inside the structure are defined, and for each of the plurality of damage patterns, the temperature distribution shape in a predetermined axial direction on the surface of the structure is associated in advance, and the surface of the structure is photographed with an infrared camera. Then, in this method for investigating a structure, the temperature distribution in a predetermined axial direction on the surface of the structure is obtained from the photographing result, the temperature distribution shape that matches the obtained temperature distribution is determined, and it is determined that the damage pattern corresponding to the determined temperature distribution shape is the damage state of the actual structure.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The temperature difference caused by internal damage of a concrete structure by the infrared thermography method is relatively small. In addition, an infrared camera having an uncooled infrared detection unit equipped with a microbolometer as a detection element (hereinafter also referred to as an "uncooled camera") has lower sensitivity and more noise generated in the captured image compared to an infrared camera having a cooled infrared detection unit (hereinafter also referred to as a "cooled camera"). Therefore, the uncooled camera may not be able to detect internal damage of a concrete structure.

[0005] This disclosure is made in view of the circumstances described above, and aims to provide an uncooled camera that can reduce noise in captured images compared to conventional cameras. [Means for solving the problem]

[0006] To achieve the above objective, the infrared camera according to the first embodiment comprises an imaging unit for imaging a subject, and an execution unit for performing image processing on the image captured by the imaging unit, wherein the imaging unit comprises an uncooled infrared detection unit for detecting infrared radiation emitted from the subject, and a black body unit covering a part of the infrared detection unit, and the execution unit comprises an acquisition unit for acquiring the captured image and noise components extracted from the captured image of the portion covered by the black body unit, and a removal unit for removing the noise components from the entire captured image.

[0007] Furthermore, in the infrared camera according to the second embodiment, the blackbody portion covers the portion of the infrared detection unit that is extended in a direction intersecting the direction in which the noise is generated.

[0008] Furthermore, in the infrared camera according to the third embodiment, the blackbody portion covers all of the infrared detection portion in the direction perpendicular to the direction in which the noise is generated, and covers only a portion of the infrared detection portion in the direction parallel to the direction in which the noise is generated.

[0009] Furthermore, the infrared camera according to the fourth embodiment further comprises a polarizer and a drive unit for rotating the polarizer, in addition to the infrared camera according to any one of the first to third embodiments, the execution unit further comprises a rotation unit and an estimation unit, the rotation unit rotates the polarizer by 180 degrees or more by controlling the drive unit, the acquisition unit acquires a plurality of captured images of the same subject while the polarizer is rotating, the estimation unit estimates a temperature estimation model expressed as a cosine function for estimating the temperature of the subject based on the detected temperature of the subject detected from the plurality of captured images and the rotation angle of the polarizer when the detected temperature is detected, and estimates the detected temperature in the state in which the infrared rays reflected from the subject are most removed, based on the temperature estimation model.

[0010] Furthermore, in the infrared camera according to the fifth embodiment, the rotating part rotates the polarizer at a constant speed, as in the infrared camera according to the fourth embodiment.

[0011] Furthermore, in the infrared camera according to the sixth embodiment, the estimation unit estimates the detected temperature in a state in which all infrared rays reflected from the subject have been removed, based on the temperature estimation model, in the infrared camera according to the fourth or fifth embodiment.

[0012] Furthermore, in the infrared camera according to the seventh embodiment, in the infrared camera according to any one of the fourth to sixth embodiments, the estimation unit estimates the value obtained by taking the difference between the offset and the absolute value of the amplitude in the temperature estimation model as the detected temperature in the state in which the infrared rays reflected from the subject are most effectively removed.

[0013] Furthermore, the image processing method according to the eighth embodiment involves a computer performing a process to obtain an image captured by an imaging unit provided in an infrared camera, and noise components extracted from the image captured by an uncooled infrared detection unit provided in the imaging unit that detects infrared radiation emitted from a subject, specifically the portion covered by the blackbody, and then removing the noise components from the entire image.

[0014] Furthermore, the image processing program according to the ninth embodiment obtains an image captured by an imaging unit provided in an infrared camera, and noise components extracted from the image captured by an uncooled infrared detection unit provided in the imaging unit that detects infrared radiation emitted from a subject, specifically the portion covered by the blackbody, and causes a computer to perform a process to remove the noise components from the entire image. [Effects of the Invention]

[0015] According to this disclosure, noise in captured images can be reduced compared to conventional methods. [Brief explanation of the drawing]

[0016] [Figure 1] This is a schematic diagram showing an example of the configuration of an infrared camera according to the first embodiment. [Figure 2] This is a block diagram showing an example of the main components of the electrical system of an infrared camera according to the first embodiment. [Figure 3] This is a schematic diagram showing an example of the configuration of the infrared detection unit according to the embodiment. [Figure 4] This block diagram shows an example of the hardware configuration of the execution unit according to the embodiment. [Figure 5] This is a functional block diagram showing an example of an execution unit according to the first embodiment. [Figure 6] This is a flowchart showing an example of image processing according to the first embodiment. [Figure 7] This is a schematic diagram illustrating an example of image processing according to the first embodiment. [Figure 8] This is an example of an image captured by changing only the exposure time without performing any image processing in the infrared camera according to the first embodiment. [Figure 9] This graph shows an example of the detected temperature when imaging a subject at a constant temperature using a cooled camera and an infrared camera according to the first embodiment. [Figure 10] This graph shows an example of the standard deviation of the evaluation domain relative to the average number of sheets. [Figure 11] A graph showing an example of the detected temperature when imaging a subject at a constant temperature using a cooled camera and an infrared camera according to the first embodiment. [Figure 12] An example of an imaging image captured using a cooled camera and an infrared camera according to the first embodiment. [Figure 13] A schematic diagram showing an example of the configuration of an infrared camera according to the second embodiment. [Figure 14] A block diagram showing an example of the main configuration of the electrical system of an infrared camera according to the second embodiment. [Figure 15] A schematic diagram showing an example of a state in which a polarizer according to the second embodiment is rotated. [Figure 16] An example of a schematic diagram for explaining the function of a polarizer according to the second embodiment. [Figure 17] A graph showing an example of the detected temperature of a subject with respect to the rotation angle of a polarizer according to the second embodiment. [Figure 18] A functional block diagram showing an example of an execution unit according to the second embodiment. [Figure 19] A flowchart showing an example of image processing according to the second embodiment. [Figure 20] A schematic diagram showing an example of a structure in which a specimen according to the second embodiment is incorporated. [Figure 21] An example of an imaging image captured using a cooled camera and an infrared camera according to the second embodiment. [Figure 22] An example of an imaging image captured using an infrared camera according to the first embodiment and an infrared camera according to the second embodiment for a subject including reflected light.

Embodiments for Carrying Out the Invention

[0017] [First Embodiment] Hereinafter, an example of an embodiment of this disclosure will be described with reference to the drawings. In each drawing, identical or equivalent components and parts are given the same reference numerals. Also, the dimensional ratios in the drawings are exaggerated for illustrative purposes and may differ from the actual ratios.

[0018] As shown in Figure 1, the infrared camera 10 according to this embodiment includes an entrance port 14 for injecting infrared light into the infrared camera 10, a lens 16, an infrared detection unit 18, and a blackbody unit 18A. The infrared camera 10 according to this embodiment is a fixed-type infrared camera. However, it is not limited to this example. The infrared camera 10 may be a non-fixed infrared camera.

[0019] The infrared detection unit 18 detects infrared radiation emitted from the subject. The infrared detection unit 18 according to this embodiment is uncooled and is composed of a microbolometer. Infrared radiation is a type of light and is classified into a wavelength range longer than visible light (for example, wavelengths from 360 nm to 830 nm) that humans can see. The wavelength detected by the infrared camera 10 according to this embodiment is from 8 μm to 14 μm. While a wavelength detected by the infrared camera 10 from 3 μm to 5 μm has the advantage of being less affected by reflected light, it has the disadvantage of being affected by sunlight, making it difficult to apply to daytime surveys.

[0020] The blackbody section 18A covers a portion of the infrared detection section 18.

[0021] Next, the main electrical components of the infrared camera 10 according to this embodiment will be described. As shown in Figure 2, the infrared camera 10 according to this embodiment has an imaging unit 30, an execution unit 32, a display unit 35, and a recording unit 36.

[0022] The imaging unit 30 is equipped with an infrared detection unit 18. The imaging unit 30 inputs the data of the image, which is a thermal image detected by the infrared detection unit 18, to the execution unit 32.

[0023] The execution unit 32 is configured to perform image processing on the image captured by the imaging unit 30. Specifically, the execution unit 32 reads the image processing program 300, which will be described later, and performs the image processing described later.

[0024] The display unit 35 receives an image signal from the execution unit 32 and displays the image captured by the imaging unit 30, as well as the captured image after image processing described later has been performed. The recording unit 36 ​​receives an image signal from the execution unit 32 and records the image captured by the imaging unit 30, as well as the captured image after image processing described later has been performed.

[0025] Next, the configuration of the infrared detection unit 18 will be described. Uncooled infrared detection units 18 tend to generate vertical noise in the captured image. Therefore, as shown in Figure 3, the infrared detection unit 18 according to this embodiment is equipped with a blackbody section 18A at the upper and lower ends, which are portions extended in a direction intersecting the direction in which noise is generated (vertical direction). Specifically, the blackbody section 18A according to this embodiment covers the entire infrared detection unit 18 in the direction perpendicular to the direction in which noise is generated, and covers only a portion of the direction parallel to the direction in which noise is generated. As a result, infrared light can be prevented from entering the infrared detection unit 18 from the upper and lower ends, so that the infrared camera 10 can extract the vertical noise component.

[0026] However, this is not the only example. For example, the infrared detection unit 18 may have a blackbody section 18A only at its upper end or only at its lower end. Alternatively, the infrared detection unit 18 may have multiple blackbody sections 18A in addition to the upper and lower ends. Furthermore, if the infrared detection unit 18 tends to generate lateral noise in the captured image, the infrared camera 10 can extract the lateral noise component by providing a blackbody section 18A at at least one of the left and right ends.

[0027] Next, the hardware configuration of the execution unit 32 will be described. As shown in Figure 4, the execution unit 32 has a CPU (Central Processing Unit) 32A, a ROM (Read Only Memory) 32B, a RAM (Random Access Memory) 32C, and a communication I / F (Interface) 32D. Each component is connected to the others via a bus 32E so that they can communicate with each other.

[0028] The CPU 32A is a central processing unit that executes various programs and controls various components. Specifically, the CPU 32A reads a program from the ROM 32B and executes the program using the RAM 32C as a working area. The CPU 32A controls each of the above components and performs various calculations according to the program stored in the ROM 32B. In this embodiment, the image processing program 300 is stored in the ROM 32B.

[0029] ROM32B stores various programs and data. RAM32C temporarily stores programs or data as a working area.

[0030] The communication I / F32D is an interface for communicating with other components such as the imaging unit 30, the display unit 35, and the recording unit 36.

[0031] Next, the functional configuration of the execution unit 32 will be described.

[0032] As shown in Figure 5, the execution unit 32 has an acquisition unit 322, a removal unit 324, and a display unit 328 as its functional configuration. Each functional configuration is realized when the CPU 32A reads and executes the image processing program 300 stored in the ROM 32B.

[0033] The acquisition unit 322 acquires an image from the imaging unit 30. The acquisition unit 322 also acquires noise components (hereinafter referred to as "noise components") from the image captured of the blackbody unit 18A. Specifically, the acquisition unit 322 acquires the value obtained by subtracting the average detected temperature of the blackbody unit 18A from the detected temperature of the blackbody unit 18A as the noise component.

[0034] The removal unit 324 removes the noise components acquired by the acquisition unit 322 from the entire image captured by the acquisition unit 322.

[0035] The display unit 328 displays the captured image, from which the noise component has been removed by the removal unit 324, on the display unit 35.

[0036] Next, the operation of the execution unit 32 will be explained.

[0037] Figure 6 is a flowchart showing the image processing flow by the execution unit 32 according to this embodiment. The CPU 32A reads the image processing program 300 from the ROM 32B, loads it into the RAM 32C, and executes it to perform image processing.

[0038] In step S100 of Figure 6, the CPU 32A acquires an image from the imaging unit 30. Hereinafter, the image acquired by the CPU 32A in step S100 will be referred to as the acquired image.

[0039] In step S102, the CPU 32A acquires noise components from the acquired image, which captures the portion covered by the blackbody 18A. Hereinafter, the noise components acquired by the CPU 32A will be referred to as acquired noise.

[0040] In step S104, the CPU 32A removes acquisition noise from the entire acquired image.

[0041] In step S106, the CPU 32A displays the noise-removed image, which is the acquired image from which the noise acquired in step S104 has been removed, on the display unit 35, and terminates this image processing.

[0042] Figure 7 is a schematic diagram illustrating the image processing according to this embodiment. In Figure 7, the left figure shows an example of the acquired image, the center figure shows an example of the acquired noise, and the right figure shows an example of the noise-removed image. The CPU 32A extracts noise from the portion of the acquired image covered by the blackbody 18A. In this embodiment, since the noise appears to extend in the vertical direction, the noise in the entire acquired image can be obtained by extending (interpolating) the noise extracted from the portion covered by the blackbody 18A. This is the acquired noise. The CPU 32A performs difference processing by removing the acquired noise from the acquired image and displays the noise-removed image on the display unit 35. The right figure in Figure 7 shows a reduction in vertical noise compared to the left figure in Figure 7.

[0043] Next, we will explain processes other than image processing that reduce noise generated in captured images.

[0044] First, the process of extending the exposure time will be explained. Figure 8 shows images acquired by changing only the exposure time without performing image processing in the infrared camera 10 according to this embodiment. Figure 8(A) shows an image acquired with a standard exposure time of 43 μm, and Figure 8(B) shows an image acquired with an exposure time of 100 μm. The image shown in Figure 8(B) has reduced vertical noise compared to the image shown in Figure 8(A). In addition, the standard deviation of the detected temperature at the evaluation line shown by the white line in Figure 8 was 0.312 in Figure 8(A) and 0.096 in Figure 8(B). Therefore, extending the exposure time is effective in reducing noise.

[0045] Figure 9 shows the detected temperature when imaging a subject at a constant temperature using a cooled camera and the infrared camera 10 according to this embodiment. Although cooled cameras can reduce noise due to their higher temperature resolution compared to uncooled cameras, they have the drawback of being more expensive than uncooled cameras. The top graph in Figure 9 shows the detected temperature when imaging is performed by a cooled camera. The second graph from the top in Figure 9 shows the detected temperature when imaging is performed by the infrared camera 10 according to this embodiment with an exposure time of 100 μm and no image processing is performed. The bottom graph in Figure 9 shows the detected temperature when imaging is performed by the infrared camera 10 according to this embodiment with an exposure time of 43 μm and no image processing is performed. From Figure 9, it can be seen that the infrared camera 10 according to this embodiment, without image processing, cannot reduce noise to the same extent as the cooled camera even when the exposure time is extended.

[0046] Next, the process of averaging multiple captured images (hereinafter referred to as "averaging process") will be explained. Figure 10 shows the standard deviation of the detected temperature at the evaluation line (hereinafter referred to as "standard deviation of the evaluation region") for the number of captured images that have undergone averaging without image processing in the infrared camera 10 according to this embodiment (hereinafter referred to as "number of averaged images"). In Figure 10, the standard deviation of the evaluation region when the exposure time is 43 μm is plotted as a circle, and the standard deviation of the evaluation region when the exposure time is 100 μm is plotted as a triangle. Also in Figure 10, the standard deviation of the evaluation region when the exposure time is 150 μm is plotted as a rectangle, and the standard deviation of the evaluation region when the exposure time is 200 μm is plotted as an X. From Figure 10, it can be seen that, regardless of the exposure time, the standard deviation of the evaluation region decreases as the number of averaged images increases. However, when the number of averaged images is 10 or more, the reduction in the standard deviation of the evaluation region by increasing the number of averaged images by one is smaller than when the number of averaged images is less than 10. Therefore, it can be seen that even if the number of images averaged is increased to 10 or more, the noise reduction effect remains the same as when the averaged number of images is 10.

[0047] Next, the effect of noise reduction through image processing and averaging will be explained. Figure 11 shows the detected temperature when an object at a constant temperature is imaged using a cooled camera and the infrared camera 10 according to this embodiment. The top graph in Figure 11 shows the detected temperature when the cooled camera takes an image. The second graph from the top in Figure 11 shows the detected temperature when image processing and averaging are performed on the infrared camera 10 according to this embodiment with an exposure time of 100 μm. The bottom graph in Figure 11 shows the detected temperature when only averaging is performed on the infrared camera 10 according to this embodiment with an exposure time of 100 μm. From Figure 11, it can be seen that even if the exposure time is extended and image processing and averaging are performed on the infrared camera 10 according to this embodiment (i.e., an infrared camera with an uncooled infrared detection unit), the standard deviation of the detected temperature cannot be reduced to the same extent as with the cooled camera.

[0048] Figure 12 shows images captured using a cooled camera and the infrared camera 10 according to this embodiment. Figure 12(A) shows an image captured using the infrared camera 10 according to this embodiment with an exposure time of 100 μm and only averaging processing performed. Figure 12(B) shows an image captured using the infrared camera 10 according to this embodiment with an exposure time of 100 μm and both image processing and averaging processing performed. Figure 12(C) shows an image captured using a cooled camera. The standard deviation of the detected temperature at the evaluation line indicated by the white line in Figure 12 was 0.047 in Figure 12(A), 0.045 in Figure 12(B), and 0.028 in Figure 12(C). Therefore, it can be seen that even if the exposure time is extended and image processing and averaging processing are performed with the infrared camera 10 according to this embodiment, the standard deviation of the detected temperature cannot be reduced to the same extent as with the cooled camera. On the other hand, the image captured in Figure 12(B) shows reduced vertical noise compared to the image captured in Figure 12(A). Furthermore, the captured image shown in Figure 12(B) shows no change in vertical noise compared to the captured image shown in Figure 12(C). Therefore, it can be seen that by extending the exposure time and performing image processing and averaging in the infrared camera 10 according to this embodiment, the noise can be reduced to a level comparable to that of a cooled camera.

[0049] As described above, the infrared camera 10 according to the first embodiment includes an imaging unit 30 for imaging a subject and an execution unit 32 for performing image processing on the captured image. The imaging unit 30 includes an uncooled infrared detection unit 18 for detecting infrared radiation emitted from the subject and a blackbody unit 18A that covers a part of the infrared detection unit 18. The execution unit 32 includes an acquisition unit 322 for acquiring noise components extracted from the captured image and the captured image of the part covered by the blackbody unit 18A, and a removal unit 324 for removing noise components from the entire captured image. As a result, the uncooled camera, which is less expensive than a cooled camera, can reduce noise in the captured image compared to conventional cameras.

[0050] Furthermore, since the infrared camera 10 according to this embodiment detects wavelengths from 8 μm to 14 μm, the infrared camera 10 can be used day or night.

[0051] In the first embodiment, the infrared camera 10 included a blackbody section 18A that covered a part of the infrared detection section 18. However, the system is not limited to this example. The infrared camera 10 according to the first embodiment does not need to include a blackbody section 18A. In this case, the acquisition section 322 creates a blurred image from the captured image using an averaging filter, and acquires noise components by subtracting the blurred image from the captured image. The removal section 324 then removes the noise components acquired from the entire captured image.

[0052] [Second Embodiment] In the first embodiment, the infrared camera 10 did not have a polarizer. In this embodiment, the infrared camera 10 removes infrared light reflected from the subject (hereinafter referred to as "reflected light") by utilizing the fact that measuring temperature by rotating a polarizer approximates a cosine wave. The differences from the first embodiment will be explained below.

[0053] As shown in Figure 13, the infrared camera 10 according to this embodiment differs from the first embodiment in that it has a polarizer 20, a support plate 22, and a motor 24.

[0054] The polarizer 20 is an optical element that transmits only light vibrating in one direction and blocks light vibrating in any other direction. As shown in Figure 15, the polarizer 20 has a fine vertical wire grid 20A arranged along a predetermined direction. Multiple wire grids 20A are arranged in parallel.

[0055] The support plate 22 is, for example, disc-shaped and rotates while supporting the polarizer 20.

[0056] The motor 24, acting as the drive unit, rotates the polarizer 20 by driving the outer circumference of the disc-shaped support plate 22.

[0057] Unlike the infrared camera 10 in the first embodiment, the infrared camera 10 in this embodiment is not limited to an uncooled camera. In other words, a cooled camera may be used as the infrared camera 10 in this embodiment.

[0058] Next, the main electrical components of the infrared camera 10 according to this embodiment will be described. As shown in Figure 14, the infrared camera 10 according to this embodiment differs from the first embodiment in that it has a motor 24.

[0059] The execution unit 32 rotates the polarizer 20 supported by the support plate 22 by controlling the motor 24.

[0060] Figure 15 shows the polarizer 20 in a rotated state. Figure 15(A) shows the polarizer 20 with a rotation angle of 0 degrees, and this position of the polarizer 20 is the reference position. Figure 15(B) shows the polarizer 20 with a rotation angle of 45 degrees relative to the reference position. Figure 15(C) shows the polarizer 20 with a rotation angle of 90 degrees relative to the reference position. Figure 15(D) shows the polarizer 20 with a rotation angle of 135 degrees relative to the reference position. When the rotation angle is 0 degrees, the amplitude direction of the reflected light (hereinafter referred to as the "reflection amplitude direction") is perpendicular to the direction of the wire grid 20A, so the polarizer 20 transmits the reflected light. On the other hand, when the rotation angle is 90 degrees, the reflection amplitude direction is parallel to the direction of the wire grid 20A, so the polarizer 20 removes the reflected light.

[0061] Figure 16 is a schematic diagram illustrating the function of the polarizer 20. In Figure 16, the rotation angle of the polarizer 20 is 0 degrees. As shown in Figure 16, the polarizer 20 transmits the reflected light W because the direction of the reflection amplitude is perpendicular to the direction of the wire grid 20A.

[0062] Figure 17 shows an example of the detected temperature of an object as a function of the rotation angle of the polarizer 20. The circles in Figure 17 indicate the detected temperature from multiple images captured with the polarizer 20 rotated 360 degrees. The cosine wave in Figure 17 represents the temperature estimation model used to estimate the detected temperature. The temperature estimation model is a cosine function expressed by equation (1). In equation (1), Yn is the detected temperature, A is the amplitude, δ is the phase difference, θ is the rotation angle of the polarizer 20, n is the frequency, and B is the offset.

[0063]

number

[0064] Figure 17 shows that when the rotation angle is 0 degrees, 180 degrees, and 360 degrees, the polarizer 20 transmits the most reflected light, resulting in the highest detected temperature. When the rotation angle is 90 degrees and 270 degrees, the polarizer 20 removes reflected light, resulting in the lowest detected temperature. Therefore, the infrared camera 10 according to this embodiment estimates the detected temperature when the rotation angle is 90 degrees and 270 degrees, i.e., when the most reflected light is removed, by taking the difference in the absolute value of the amplitude A from the offset B in the temperature estimation model. Alternatively, the infrared camera 10 may estimate the detected temperature when the most reflected light is removed by substituting 90 degrees or 270 degrees for θ in the temperature estimation model. Hereinafter, the estimated value of the detected temperature when the most reflected light is removed will be referred to as the "estimated temperature". Note that if the polarizer 20 can remove all reflected light, the detected temperature will be the detected temperature when all reflected light is removed.

[0065] Next, the functional configuration of the execution unit 32 will be described.

[0066] As shown in Figure 18, the execution unit 32 differs from the first embodiment in that it includes a rotating unit 320 and an estimation unit 326.

[0067] The rotating unit 320 controls the motor 24 to rotate the polarizer 20, which is supported by the support plate 22, by 180 degrees or more. In this embodiment, the rotating unit 320 controls the motor 24 so that the polarizer 20 rotates in the circumferential direction at a constant speed.

[0068] The acquisition unit 322 differs from the first embodiment in that it acquires multiple captured images of the same subject while the polarizer 20 is rotating. In this embodiment, the acquisition unit 322 acquires an captured image each time the polarizer 20 rotates by a predetermined angle (e.g., 40 degrees). However, it is not limited to this example. For example, if the polarizer 20 is rotating at a constant speed, the acquisition unit 322 may acquire an captured image each time a predetermined amount of time (e.g., 0.1 seconds) has elapsed, regardless of the angle of the polarizer 20.

[0069] Furthermore, the acquisition unit 322 differs from the first embodiment in that it acquires the detected temperature of the subject and the rotation angle of the polarizer 20 when the detected temperature is detected, for each pixel, from multiple captured images.

[0070] The estimation unit 326 estimates a temperature estimation model based on the detected temperature of the subject acquired by the acquisition unit 322 and the rotation angle of the polarizer 20 when the detected temperature is detected, and estimates the temperature based on the temperature estimation model.

[0071] Next, the operation of the execution unit 32 will be explained.

[0072] Figure 19 is a flowchart showing the image processing flow by the execution unit 32 according to this embodiment. The CPU 32A reads the image processing program 300 from the ROM 32B, loads it into the RAM 32C, and executes it to perform image processing.

[0073] In step S200 of Figure 19, the CPU 32A controls the motor 24 to rotate the polarizer 20 by more than 180 degrees.

[0074] In step S202, the CPU 32A acquires multiple images of the same subject while the polarizer 20 is rotating. Hereinafter, the images acquired by the CPU 32A in step S202 will be referred to as rotated images.

[0075] In step S204, the CPU 32A acquires noise from the rotational image captured of the portion covered by the blackbody 18A.

[0076] In step S206, CPU32A removes acquired noise from the entirety of all rotated images.

[0077] In step S208, the CPU 32A obtains the detected temperature for each pixel and the rotation angle of the polarizer 20 when that temperature is detected from all the acquired rotation images.

[0078] In step S210, the CPU 32A estimates a temperature estimation model by performing nonlinear regression based on the detected temperature obtained in step S208 and the rotation angle of the polarizer 20.

[0079] In step S212, CPU32A estimates the temperature by taking the difference in the absolute value of the amplitude A from the offset B in the temperature estimation model estimated in step S210.

[0080] In step S214, the CPU 32A determines whether or not it has estimated the temperature for all pixels of the rotated image. If the CPU 32A has estimated the temperature for all pixels of the rotated image (step S214: YES), it proceeds to step S216. On the other hand, if the CPU 32A has not estimated the temperature for all pixels of the rotated image (step S214: NO), it returns to step S210.

[0081] In step S216, the CPU 32A displays the thermal image of the estimated temperature as an image captured by the imaging unit 30 on the display unit 35, and terminates the image processing.

[0082] Next, the noise reduction effect achieved by performing the image processing according to this embodiment will be explained. Figure 20 shows a structure S in which the test specimen T is embedded. In Figure 20, the test specimen T is shown by a dotted line and the structure S is shown by a solid line. Figure 20(A) shows the state in which the test specimen T is embedded at a cover thickness of 2 cm, Figure 20(B) shows the state in which the cover thickness is 3 cm, and Figure 20(C) shows the state in which the cover thickness is 4 cm. The width and height of the test specimen T are 150 mm.

[0083] Figure 21 shows images captured when the structure S shown in Figure 20 is imaged using a cooled camera and an infrared camera 10 according to this embodiment. Figure 21(A) shows an image captured by the infrared camera 10 without performing the image processing according to this embodiment. Figure 21(B) shows an image captured by the infrared camera 10 after performing the image processing according to this embodiment. The initial values ​​of the temperature estimation model estimated in the image processing are set as follows: frequency n is twice the rotation speed of the polarizer 20, phase difference δ is 0, amplitude A is the difference of the detected temperature Yn, and offset B is the average value of the detected temperature. Figure 21(C) shows an image captured by the cooled camera.

[0084] The image shown in Figure 21(A) is noisier than the images shown in Figures 21(B) and 21(C). Furthermore, in the left image of Figure 21(A), the temperature difference between the test specimen T embedded at a position with a cover thickness of 4 cm and the structure S is detected as approximately 0.1°C. In other words, the uncooled camera that does not perform image processing according to this embodiment cannot detect the test specimen T embedded at a position with a cover thickness of 4 cm due to the high level of noise.

[0085] On the other hand, the captured image shown in Figure 21(B) has less noise compared to the captured image shown in Figure 21(A), and generates only about the same amount of noise as the captured image shown in Figure 21(C). Furthermore, in Figures 21(B) and 21(C), the temperature difference between the test specimen T, which is embedded at a position with a cover thickness of 4 cm, and the structure S is detected as approximately 3.0°C. Therefore, it can be seen that even with an uncooled camera, noise can be removed in the same way as with a cooled camera by performing the image processing according to this embodiment.

[0086] Figure 22 also shows images of voids, which are internal structures within a bridge, captured using the infrared camera according to the first embodiment and the infrared camera according to the second embodiment. Figure 22(A) shows an image captured with noise removed using the infrared camera according to the first embodiment. Figure 22(B) shows an image captured with noise removed using the infrared camera 10 according to this embodiment, and with reflected light removed by rotating the polarizer 20. In other words, the difference between Figure 22(A) and Figure 22(B) is whether or not the process of removing reflected light by rotating the polarizer 20 is performed. Also, the area enclosed by the white dotted line in Figure 22 is the area where there is a void (hereinafter referred to as the "void area"). In Figure 21(A), the void area is overlapped with reflected light, and the temperature difference caused by the void cannot be confirmed. On the other hand, in Figure 22(B), the reflected light has been removed from the void area, so the temperature difference caused by the void can be confirmed. Therefore, by performing image processing in the infrared camera 10 according to this embodiment, it becomes possible to detect temperature differences in the internal structure and damage of the bridge even when reflected light is generated.

[0087] As described above, the infrared camera 10 according to the second embodiment further comprises a polarizer 20 and a motor 24 for rotating the polarizer 20. Furthermore, the execution unit 32 according to the second embodiment further comprises a rotation unit 320 and an estimation unit 326. The rotation unit 320 rotates the polarizer 20 by 180 degrees or more by controlling the motor 24. The acquisition unit 322 acquires multiple rotation images of the same subject while the polarizer 20 is rotating. The estimation unit 326 estimates an estimation model based on the detected temperature of the subject detected from the multiple rotation images and the rotation angle of the polarizer 20 when the detected temperature is detected, and estimates the temperature based on the temperature estimation model. As a result, reflected light can be reduced compared to the conventional method. Therefore, for example, in damage investigation of concrete structures and exterior wall diagnosis of buildings, it is possible to reduce the possibility that reflected light from the surrounding thermal environment of the subject being investigated will be captured by the infrared camera 10, and that the temperature difference due to such reflected light will be mistakenly detected as internal damage. Furthermore, the reflection of light can overlap with the internal damage, reducing the possibility of overlooking the damage.

[0088] This disclosure is not limited to the embodiments described above, and various modifications and applications are possible without departing from the spirit of the invention.

[0089] For example, the execution unit 32 in each of the above embodiments was built into the infrared camera 10. However, the example is not limited to this. For example, the execution unit 32 may be configured separately from the infrared camera 10.

[0090] Furthermore, in each of the above embodiments, the blackbody portion 18A is provided to cover two locations of the infrared detection unit 18, namely the upper and lower ends, which are portions extended in a direction intersecting the direction in which noise is generated (vertical direction). However, it is not limited to this. The blackbody portion 18A may be provided to cover the infrared detection unit 18 in one or three or more locations. Also, the blackbody portion 18A does not necessarily have to cover the ends of the infrared detection unit 18. As long as it extends in a direction intersecting the direction in which noise is generated, it may be provided to cover the infrared detection unit 18 at a location away from the ends. In addition, although the above embodiments described the case in which the noise extends in the vertical direction of the drawing, if the noise extends in the horizontal direction, the blackbody portion 18A may be provided to extend in the direction intersecting the direction in which the noise extends, i.e., in the vertical direction.

[0091] Furthermore, the various processes that the CPU 32A reads and executes in each of the above embodiments may be executed by various processors other than the CPU. Examples of such processors include PLDs (Programmable Logic Devices) such as FPGAs (Field-Programmable Gate Arrays) whose circuit configuration can be changed after manufacturing, and dedicated electrical circuits that are processors with circuit configurations specifically designed to execute specific processes, such as ASICs (Application Specific Integrated Circuits). In addition, the above various processes may be executed by one of these various processors, or by a combination of two or more processors of the same or different types (for example, multiple FPGAs, and a combination of a CPU and an FPGA). More specifically, the hardware structure of these various processors is an electrical circuit that combines circuit elements such as semiconductor elements.

[0092] Furthermore, although the above embodiments describe an embodiment in which the image processing program 300 is pre-stored (installed) in the ROM 32B, the invention is not limited thereto. The program may be provided in the form of a recording medium such as a CD-ROM (Compact Disc Read Only Memory), DVD-ROM (Digital Versatile Disc Read Only Memory), or USB (Universal Serial Bus) memory. Alternatively, the program may be provided in the form of a download from an external device via a network.

[0093] Furthermore, the processing flow described in each of the above embodiments is merely an example, and unnecessary steps may be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0094] Furthermore, the configurations of the infrared camera 10 described in each of the above embodiments are examples and may be modified as needed without departing from the main point. [Explanation of Symbols]

[0095] 10 Infrared Cameras 18 Infrared detection unit 18A Blackbody 20 polarizers 20A Wire Grid 24 motors 30 Imaging Unit 32 Execution Department 300 Image Processing Programs 320 Rotating part 322 Acquisition Department 324 Removal section 326 Estimation Department

Claims

1. The imaging unit captures images of the subject, The system includes an execution unit for performing image processing on the captured image captured by the imaging unit, The imaging unit is An uncooled infrared detection unit for detecting infrared radiation emitted from the subject, It comprises a black body portion that covers a part of the infrared detection portion, The execution unit is, An acquisition unit that acquires the captured image and noise components extracted from the captured image in which the portion covered by the blackbody is captured, The system includes a removal unit that removes the noise component from the entire captured image, The blackbody portion covers the portion of the infrared detection unit that is extended in a direction intersecting the direction in which the noise is generated. Infrared camera.

2. The blackbody portion covers all of the infrared detection portion in the direction perpendicular to the direction in which the noise is generated, and covers only a portion of the portion parallel to the direction in which the noise is generated. The infrared camera according to claim 1.

3. The device further comprises a polarizer and a drive unit for rotating the polarizer, The execution unit further comprises a rotation unit and an estimation unit, The rotating part rotates the polarizer by 180 degrees or more by controlling the drive unit. The acquisition unit acquires a plurality of captured images of the same subject while the polarizer is rotating. The estimation unit estimates a temperature estimation model represented by a cosine function that estimates the temperature of the subject based on the detected temperature of the subject detected from the plurality of captured images and the rotation angle of the polarizer when the detected temperature is detected, and estimates the detected temperature in the state in which the infrared rays reflected from the subject are most effectively removed, based on the temperature estimation model. The infrared camera according to claim 1.

4. The rotating part rotates the polarizer at a constant speed. The infrared camera according to claim 3.

5. The estimation unit estimates the detected temperature in a state where all infrared radiation reflected from the subject has been removed, based on the temperature estimation model. The infrared camera according to claim 3.

6. The estimation unit estimates the value obtained by taking the difference between the offset and the absolute value of the amplitude in the temperature estimation model as the detected temperature in the state in which the infrared radiation reflected from the subject is most effectively removed. The infrared camera according to claim 3.

Images