JP2000258584A - On-site inspection device - Google Patents

Abstract

(57) [Summary] [PROBLEMS] To provide an on-site inspection device capable of accurately measuring the temperature of an inspection target device regardless of the state of the inspection target device and automating the inspection. SOLUTION: A semi-transmissive mirror 11 of an optical device 9 transmits a part of light from an inspection object 1, enters a digital camera 13 of an imaging device 10, and reflects the remaining light. The light reflected by the semi-transmissive mirror 11 is totally reflected by the total reflection mirror 12 of the optical device 4 and is input to the infrared camera 14 of the imaging device 10. In the digital camera 13, the inspection object 1
Is photographed, and the infrared camera 14 photographs the temperature distribution on the surface of the inspection object 1. The inspection processing device 4 inspects the inspection target 1 for abnormalities based on the video data of the inspection target device 1 obtained by the digital camera 13 and the video data of the inspection target device 1 obtained by the infrared camera 14.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an on-site inspection device for performing image processing on the state of equipment in a plant and automatically detecting an abnormality thereof.

[0002]

2. Description of the Related Art Recently, in a power generation plant or an industrial plant, a site inspection device has been developed which performs a monitoring operation of a site device of the plant by image processing using a computer. In this method, the status of field devices is displayed on a display device located away from the site using image processing of a computer, so that the field devices can be monitored visually by humans. In particular, there is a strong demand for the introduction of such an on-site inspection device in a radiation control area or a high risk area in an emergency at a nuclear power plant.

FIG. 8 shows such a conventional on-site inspection apparatus. The state of the inspection target device 1 is input to the inspection treatment device 4 via the TV camera 2 which is a visible wavelength band imaging device and the infrared camera 3 which is an infrared wavelength band imaging device, and the state of the inspection target device 1 is checked by the inspection device 4. Be monitored. The video data input via the TV camera 2 is converted into digital data by the A / D converter 5 and compared with normal image data stored in the image database 7 by the comparison processing means 6 in advance.

That is, normal image data is stored in an image database in advance, and a comparison between the detected image data and normal image data in the image database 7 detects whether or not an abnormal state exists.

On the other hand, the infrared image data input via the infrared camera 3 is input to the monitor device 8, and the temperature distribution image of the inspection target device 1 is displayed in color to be provided to the monitoring person, and the maximum temperature value, The minimum temperature value, histogram, and histogram on the line are provided to the observer.

[0006]

However, the monitoring device 8
Since the infrared image displayed in (1) is an image of the temperature distribution, the external shape of the inspection target device 1 is difficult to understand, and it may not be clear what the user is looking at.

[0007] Since the infrared image is obtained by imaging the radiant energy of the inspection target device 1, it does not represent the actual temperature distribution. That is, it depends on the shape and surface condition of the inspection target device 1. In addition, since the detected temperature distribution image depends on the surface state and the like, if dust or oil is attached to the surface of the inspection target device 1, the image (temperature distribution) changes.

An object of the present invention is to provide an on-site inspection apparatus capable of accurately measuring the temperature of an inspection target device regardless of the state of the inspection target device and automating the inspection.

[0009]

According to a first aspect of the present invention, there is provided an on-site inspection apparatus for photographing an object to be inspected in a visible wavelength band, and an infrared wavelength band imaging for photographing a temperature distribution on a surface of the object to be inspected. An image pickup device having a device; a semi-transmissive mirror that transmits a part of light from the inspection object, enters the visible wavelength band image pickup device and reflects the remaining light, and the semi-transmissive mirror. An optical device having a total reflection mirror that totally reflects reflected light and inputs the reflected light to the infrared wavelength band imaging device; image data of the inspection target device obtained by the visible wavelength band imaging device and the infrared wavelength band An inspection processing device for inspecting an abnormality of the inspection target based on video data of the inspection target device obtained by an imaging device.

According to the on-site inspection device according to the first aspect of the present invention, the semi-transparent mirror of the optical device transmits a part of the light from the inspection object, inputs the light to the visible wavelength band imaging device of the imaging device, and leaves the remaining light. Reflects light. The light reflected by the semi-transmissive mirror is totally reflected by the total reflection mirror of the optical device and is input to the infrared wavelength band imaging device of the imaging device. Then, the visible wavelength band imaging apparatus photographs the external shape of the inspection object, and the infrared wavelength band imaging apparatus photographs the temperature distribution on the surface of the inspection object. The inspection processing device inspects the inspection target for an abnormality based on video data of the inspection target device obtained by the visible wavelength band imaging device and video data of the inspection target device obtained by the infrared wavelength band imaging device.

According to a second aspect of the present invention, an on-site inspection device is provided.
2. The inspection processing device according to claim 1, wherein the inspection processing device extracts a circular component and a line component of the inspection target device shape based on a video signal from the visible wavelength band imaging device, and the image recognition device. Comparing and comparing the circle component and the line component extracted in step 1 with the device recognition data in which the characteristics of the inspection target device are described in advance with the circle component and the line component, and specifying the inspection target device; Image processing means for converting into an image having the number of pixels in the direction from the center of gravity after the division processing into regions for each temperature based on a video signal from the wavelength band imaging device, and a feature amount obtained by the image processing means; A device abnormality recognizing means for comparing and collating with a characteristic amount of a temperature distribution of the inspection target device in a normal state.

[0012] According to the on-site inspection device according to the second aspect of the present invention, in addition to the operation of the first aspect, the inspection processing device includes a circular component of the shape of the device to be inspected based on the video signal from the visible wavelength band imaging device. And the line component, and compares the extracted circle component and the line component with the device recognition data in which the characteristics of the inspection target device are described in advance with the circle component and the line component. After the division processing into regions for each temperature based on the video signal, the image is converted into an image having the number of pixels in the direction from the center of gravity as a feature amount. Then, the characteristic amount is compared with the characteristic amount of the temperature distribution of the inspection target device in a normal state, and an abnormality of the inspection target device is checked.

According to a third aspect of the present invention, there is provided an on-site inspection device,
Imaging apparatus having a visible wavelength band imaging device for imaging an inspection object, and two first infrared wavelength band imaging devices and a second infrared wavelength band imaging device for imaging a temperature distribution on the surface of the inspection object. A first semi-transmissive mirror that transmits a part of light from the inspection object, enters the visible wavelength band imaging device, and reflects the remaining light; and a first semi-transparent mirror. A second semi-transmissive mirror that reflects a part of the reflected light, enters the first infrared wavelength band imaging device through the first filter, and transmits the remaining light, and the second semi-transparent mirror. An optical device including: a total reflection mirror that totally reflects light transmitted through the transmissive mirror and inputs the light to the second infrared wavelength band imaging device through the first filter and a second filter having a different transmission wavelength. Obtained by the visible wavelength band imaging device; Inspection processing for inspecting an abnormality of the inspection target based on video data of the inspection target device and video data of the inspection target device obtained by the first infrared wavelength band imaging device and the second infrared wavelength band imaging device. And a device.

According to the third aspect of the present invention, the first semi-transparent mirror of the optical device transmits a part of the light from the inspection object, enters the light into the visible wavelength band imaging device, and inputs the remaining light. Is reflected. The remaining part of the light reflected by the first semi-transmissive mirror is input to the first infrared wavelength band imaging device via the first filter by the second semi-transparent mirror, and the remaining light is The light passes through the second semi-transmissive mirror and is totally reflected by the total reflection mirror. The light totally reflected by the total reflection mirror is input to the second infrared wavelength band imaging device via the first filter and the second filter having a different transmission wavelength.
The visible wavelength band imaging device photographs the external shape of the inspection object, and the two first infrared wavelength band imaging devices and the second infrared wavelength band imaging device image the temperature distribution on the surface of the inspection object. The inspection processing device performs inspection based on video data of the inspection target device obtained by the visible wavelength band imaging device, and video data of the inspection target device obtained by the infrared wavelength band imaging device and the second infrared wavelength band imaging device. Check the target for abnormalities.

According to a fourth aspect of the present invention, an on-site inspection device is provided.
4. The image processing device according to claim 3, wherein the inspection processing device is configured to extract a circular component and a line component of the inspection target device shape based on a video signal from the visible wavelength band imaging device, and the image recognition device. Comparing and comparing the circle component and the line component extracted in the above with the device recognition data in which the characteristics of the inspection target device are described in advance as the circle component and the line component, and specifying the inspection target device; Calculating a temperature that does not include the influence of the emissivity of the inspection target device based on the video signal from the first infrared wavelength band imaging device and the video signal from the second infrared wavelength band imaging device; Image processing means for converting the number of pixels from the center of gravity after the division processing into an image having a feature quantity, and the feature quantity obtained by the image processing means and the feature quantity of the temperature distribution of the inspection target device in a normal state. Characterized by comprising a device abnormality recognizing means for compare match.

According to a fourth aspect of the present invention, in addition to the function of the third aspect, in the inspection processing device, the inspection processing device has a circular component of the shape of the inspection target device based on a video signal from the visible wavelength band imaging device. And a line component are extracted, and the extracted circle component and line component are compared with device recognition data in which characteristics of the device to be inspected are described in advance by using a circle component and a line component to identify the device to be inspected. Then, based on the video signal from the first infrared wavelength band imaging device and the video signal from the second infrared wavelength band imaging device, a temperature that does not include the influence of the emissivity of the inspection target device is calculated. Is converted into an image having the number of pixels in the direction from the center of gravity as a feature value after the division processing into the region of. The characteristic amount is compared with the characteristic amount of the temperature distribution of the inspection target device in the normal state, and the inspection target is inspected for abnormality.

According to a fifth aspect of the present invention, there is provided an on-site inspection device comprising:
In the invention according to any one of claims 1 to 4,
A display output device for displaying and recording an inspection result obtained by the inspection processing device is provided.

According to a fifth aspect of the present invention, in addition to the function of the first aspect, the inspection result obtained by the inspection processing device is displayed on the display output device. And record.

According to a sixth aspect of the present invention, there is provided an on-site inspection device,
The invention according to claim 1 or 3, further comprising: a moving device that mounts and moves the imaging device, the optical device, and the inspection processing device, and a control device that controls a turning angle and an elevation angle of the imaging device. It is characterized by the following.

According to a sixth aspect of the present invention, in addition to the functions of the first or third aspect of the present invention, an image pickup device, an optical device, and an inspection processing device are mounted on a moving device, and equipment to be inspected is provided. Move to near. Then, the turning angle and the elevation angle of the imaging device are controlled by the control device, and the imaging device is directed to the imaging location of the inspection target device.

According to a seventh aspect of the present invention, there is provided an on-site inspection device,
The invention according to claim 2 or 4, further comprising a setting device for setting a characteristic amount of a temperature distribution of the inspection target device in a normal state used for abnormality determination by the device abnormality recognition unit. And

According to the on-site inspection device according to the seventh aspect of the present invention, in addition to the operation of the second or fourth aspect of the present invention, the normal inspection used by the setting device for the abnormality determination by the device abnormality recognizing means. The feature value of the temperature distribution of the target device is set.

[0023]

Embodiments of the present invention will be described below. FIG. 1 is a configuration diagram of a site inspection device according to the first embodiment of the present invention.

Light from the inspection target device 1 is input to the imaging device 10 via the optical device 9. The optical device 9 includes a semi-transmissive mirror 11 and a total reflection mirror 12, and the imaging device 10 includes a digital camera 13 that is a visible wavelength band imaging device and an infrared camera 1 that is an infrared wavelength band imaging device.
4. The digital camera 13 captures an image of the inspection target device 1 with a digital signal, and can be replaced by a combination of the TV camera 2 and the A / D converter 5. The semi-transmissive mirror 11 and the total reflection mirror 12 of the optical device 9 are installed in combination so that the optical axes X of the digital camera 13 and the infrared camera 14 are the same.

The semi-transmissive mirror 11 transmits a part of the light from the inspection object 1 and transmits the light to the digital camera 13 of the imaging device 9.
And reflects the remaining light. The light transmitted through the semi-transmissive mirror 11 is input to the digital camera 13,
The light reflected by the semi-transmissive mirror 11 is totally reflected by the total reflection mirror of the optical device 9, and is reflected by the infrared camera 14 of the imaging device 10.
Is input to The digital camera 13 photographs the external shape of the inspection object 1, and the infrared camera 14 photographs the inspection object 1.
The temperature distribution on the surface of is photographed. Image data of the inspection target device 1 obtained by the digital camera 13 and the infrared camera 1
The video data of the inspection target device 1 obtained in step 4 is input to the inspection processing device 4, where the abnormality of the inspection target 1 is checked and determined.

The image recognizing means 15 of the inspection processing device 4 extracts a circular component and a linear component of the shape of the inspection target device 1 based on the video signal from the digital camera 13 and compares the extracted components with the comparison operation means 1.
6 is output. In the comparison operation means 16, the image recognition means 1
The inspection target device 1 is recognized by comparing the circle component and the line component extracted in step 5 with the device recognition data stored in the device recognition database 17 in advance.

That is, the equipment recognition database 17 previously stores equipment recognition data described by a circle component and a line component indicating the characteristics of the inspection target equipment 1, and extracts the data extracted from the image recognition means 15 and its data. The inspection target device 1 is specified by comparing and collating with the device recognition data.
The information of the inspection target device 1 specified here is output to the display output device 18.

On the other hand, a video signal from the infrared camera 14 is converted into a digital signal by the A / D converter 20 and input to the image processing means 19. The image processing unit 19 divides the image into regions for each temperature based on the video signal from the infrared camera 14, and after the processing, converts the divided region into an image having the number of pixels in the direction from the center of gravity on the image as a feature amount. The data is converted and output to the device abnormality recognition means 21. The device abnormality recognition unit 21 compares and compares the characteristic amount obtained by the image processing unit 19 with the characteristic amount of the temperature distribution of the inspection target device 1 in a normal state stored in the abnormality recognition database 22 in advance. Thereby, it is determined whether or not the inspection target device 1 has an abnormality, and the determination result is output to the display output device 18.

As described above, the inspection result obtained by the inspection processing device 4 is output to the display output device 18 and is recorded as well as displayed. As a result, it is possible to display and output the inspection results at the past time.

FIG. 2 is an explanatory diagram of the image recognition processing by the image recognition means 15. FIG. 2A is a block diagram of the image recognition unit 15, and FIG. 2B is an explanatory diagram of the image. FIG.
As shown in (a), the image G from the digital camera 13
1 is differentiated by a differentiating means 15a to extract a portion having a large luminance change. Then, the binarization processing means 15b
The line component L is extracted from the image G2 which has been subjected to the binarization process and binarized by the line component extraction unit 15c, and the circle component S is extracted by the circle component extraction unit 15d. The extracted information is input to the comparison operation means 16 and is compared with the device recognition data in the device recognition database 17 to specify the inspection target device to be photographed.

Here, in the device recognition database 17, device recognition data created by previously extracting a portion of each inspection target device 1 that can be represented by a line and a circle, that is, each inspection target device 1 includes a plurality of lines. Also, information expressed in relation to the size and position of the circle is stored.

For example, the piping image F shown in FIG.
Since it is composed of the circle component S1 to the circle component S3 and the line component L1 to the line component L6, these features are characterized and stored in advance for each inspection target device as shown in FIG. For example,
The circle component S1 and the circle component S2 have the same center and the size ratio is 1: 0.9, and the circle component S1 is connected to the line component L3 and the line component L6... And is stored only in the device recognition database 17 as device recognition data.

In the comparison operation means 16, the digital camera 1
A device in which the size relationship and the positional relationship between the line component and the circle component obtained by processing the image from No. 3 match in the device recognition database 17 is recognized as an inspection target. As described above, by using the inspection target device 1 as a database of the relationship between the feature amounts, the inspection target device 1 can be identified without being affected by the viewpoint and the distance.

On the other hand, the image from the infrared camera 14 is processed by the image processing means 19. FIG. 4 is an explanatory diagram of the image processing by the image processing means 19. FIG. 4A is a block diagram of the image processing means 19, and FIG. 4B is an explanatory diagram of the image. The image obtained by the infrared camera 14 is A / D
Since the temperature is expressed as luminance when A / D converted by the converter 20, the infrared image H is divided by the dividing means 19a according to the luminance. Thereby, the image is divided into a plurality of images for each temperature. The calculating means 19b obtains the center of gravity for each of the divided images for each temperature, and converts the boundary line into information J based on the distance from the center of gravity. The converted information J is input to the device abnormality recognizing means 21 and is compared with the abnormality recognizing data of the abnormality recognizing database 22 to determine the abnormality of the inspection target device 1 to be photographed.

The abnormality recognition database 21 previously stores abnormality recognition data obtained by converting the normal temperature distribution of each inspection target device 1 into a size (distance from the center of gravity) for a certain temperature. I have.

As shown in FIG. 4B, since the brightness of the infrared image H differs depending on the temperature, for example, a portion having a brightness equal to or lower than 200 ° C. and a portion having a brightness equal to or higher than 200 ° C. The regions are extracted, and the images H1 for each temperature are extracted.
Divide into H2. Then, the center of gravity of the extracted portion of each of the images H1 and H2 is obtained, and the boundary line is converted into information J1 and J2 based on the distance from the center of gravity. And this information J1,
J2 is compared with the abnormality recognition data in the abnormality recognition database 22, and if the distance information from the center of gravity is significantly different, the temperature of the inspection target device 1 is recognized as abnormal and an alarm is issued to the display output device 18.

Since the digital camera 13 and the infrared camera 14 use the same optical axis X, the center points coincide with each other. Therefore, the angles of view of the digital camera 13 and the infrared camera 14 are checked in advance, and how many times the size corresponds is checked. Then, enlarge and enlarge the image taken and converted
By reducing and arranging the images side by side and comparing them, the inspection target device 1 can be easily identified.

As described above, according to the first embodiment, an image for determining the external shape of the object to be inspected and an infrared image indicating the surface temperature thereof are obtained as images on the same optical axis. Inspection except for the blind spot between the cameras of the device 10 becomes possible. Also, by measuring the angle of view between the TV camera or digital camera and the infrared camera in advance,
The scale can be made the same by enlarging / reducing from the center. Therefore, when the visible image and the infrared image are compared, the inspection target device corresponding to the temperature distribution can be easily determined.

Next, a second embodiment of the present invention will be described. FIG. 5 is a configuration diagram of a site inspection device according to the second embodiment of the present invention. The second embodiment differs from the first embodiment shown in FIG. 1 in that two first infrared wavelength band imaging devices (infrared cameras) 14a are provided as imaging devices.
14b, the first filters 23 having different transmission wavelengths.
Infrared images are input via a and 23b to remove the influence of emissivity that varies depending on the surface condition of the inspection target device 1.

In FIG. 5, the first semi-transmissive mirror 11a of the optical device 9 transmits a part of the light from the inspection object 1 and inputs it to the TV camera 2, and reflects the remaining light to form the second semi-transparent mirror 11a. The light is input to the semi-transmissive mirror 11b. Second semi-transparent mirror 1
1b is reflected by the first filter 23a.
Is input to the infrared camera 14a. The light transmitted through the second semi-transmissive mirror 11b is totally reflected by the total reflection mirror 12, and is input to the second infrared camera 14b via the second filter 23b. The transmission wavelength of the second filter 23b is different from the transmission wavelength of the first filter 23a.

As described above, in the second embodiment, the optical axis X uses the two semi-transmissive mirrors 11a and 11b, and the infrared cameras 14a through the filters 23a and 23b having different transmission wavelengths. , 14b, the light from the inspection target device 1 is input. This is to remove the influence of the emissivity of the inspection target device 1 by applying the principle of the two-color thermometer since the emissivity changes depending on the surface state of the inspection target device 1 in the infrared image. That is, the spectral radiance of the object is represented by the following equation.

N′λ = f (λ) · ε (λ) · s (λ) · πC ₁ λ ^-5 [e
xp (C ₂ / λT) ^-1 ] ^-1 where f (λ) is the transmission wavelength of the filter and ε (λ) is the wavelength λ
, S (λ) is the sensitivity of the detector at wavelength λ, and λ is the filter wavelength.

Now, the wavelengths of the two filters 23a and 23b are λ1 and λ2, and the infrared cameras 14a and 14b
When the inspection target device 1 is photographed in b and the ratio of the image obtained by A / D conversion of the photographed video is obtained, it is represented by the following equation.

R = ｛f (λ ₁ ) ・ ε (λ ₁ ) ・ s (λ ₁ ) ・ πC ₁ λ ₁
^-5 [exp (C ₂ / λ ₁ T) ^-1 ] ^-1 ｝ / ｛f (λ ₂ ) ・ ε (λ ₂ ) ・ s
(λ ₂ ) · πC ₁ λ ₂ ^-5 [exp (C ₂ / λ ₂ T) ⁻¹ ] ⁻¹ ｝ where the transmission wavelength λ of the two filters 23 a and 23 b
If λ1 and λ2 are set close to each other, ε (λ ₁ ) = ε (λ ₂ ). Therefore, the above equation is represented by the following equation.

R = ｛f (λ ₁ ) ・ s (λ ₁ ) ・ πC ₁ λ ₁ ^-5 [exp
(C ₂ / λ ₁ T) ⁻¹ ] ⁻¹ ｝ / ｛f (λ ₂ ) · s (λ ₂ ) · πC ₁ λ ₂
^-5 [exp (C ₂ / λ ₂ T) ⁻¹ ] ⁻¹ ｝ This makes it possible to measure the temperature T of the inspection target device 1 that does not include the influence of the emissivity of the inspection target device 1.

The calculation for removing the effect of the emissivity is performed by the image processing means 19, and furthermore, the image processing means 19
Then, the abnormality is determined by comparing the thermal image obtained by this calculation with the abnormality recognition data stored in the abnormality recognition database 22 in advance. On the other hand, the video of the TV camera 2 is A / D converted and the digital camera 1 of the first embodiment is converted.
3 and processed by the comparison operation means 7 to specify the inspection target device 1. Then, the result is output to the display output device 18.

As described above, according to the second embodiment, the image of the external shape of the inspection target device 1 and the infrared image indicating the surface temperature are obtained as images on the same optical axis. The effect of emissivity is eliminated by eliminating the blind spot between the ten cameras and by determining the intensity ratio using two types of filters of the center wavelength, so that accurate temperature measurement can be performed.

FIG. 6 shows a case where the imaging device 10, the optical device 9, and the inspection processing device 4 shown in the first embodiment and the second embodiment are mounted on the moving device 23, and the inspection target equipment at the site of the plant. It can be moved to 1. This makes it possible to expand the inspection targets.

The optical device 9 and the imaging device 10 are mounted on the upper part of the moving device 23. In FIG. 6, illustration of the optical device 9 is omitted. An elevation mechanism 24 is attached to the imaging device 10, and the turning angle and the elevation angle of the imaging device 10 are adjusted by the elevation mechanism 24. This raising mechanism 24
Is controlled by the control device 25.

Accordingly, the imaging device 10 is moved to the vicinity of the inspection target device 1 by the moving device 23, and the imaging device 10 is adjusted to the position where the inspection image of the inspection target device 1 is photographed by the elevating mechanism 24.

As described above, the mobile device 23 that can be remotely controlled is used.
Since the imaging device 10, the optical device 9, and the inspection processing device 4 are mounted on the inspection device 1 to move between the inspection target devices 1, it is also possible to inspect a narrow portion or the like where humans cannot approach.

Next, a third embodiment of the present invention will be described. 7A and 7B are explanatory diagrams of a site inspection device according to a third embodiment of the present invention. FIG. 7A is a partial configuration diagram, and FIG. 7B is a flowchart showing the operation.

The third embodiment is similar to the first embodiment shown in FIG.
5 and the second embodiment shown in FIG.
A setting device 26 is additionally provided for setting a characteristic amount of a temperature distribution of a device to be inspected in a normal state, which is used for abnormality determination by the device abnormality recognizing means 21 of the inspection processing device 4.

As shown in FIG. 7A, the monitor sets the abnormality recognition data in the abnormality recognition database 22 using the setting device 26. For example, a change in the operation state of the plant equipment may require that the abnormality recognition data stored in advance in the abnormality recognition database 22 be changed. Even if the surface temperature of the plant equipment should be determined to be abnormal in a certain operation state, it may be determined to be normal if the operation state of the plant equipment is changed. In such a case, the monitoring person sets and adds new abnormality recognition data to the abnormality recognition database 22 using the setting device 26.

FIG. 7B is a flowchart showing the operation in that case. As described above, the device abnormality recognizing unit 21 compares the temperature distribution information of the infrared image obtained from the image processing unit 19 with the abnormality recognizing data of the abnormality recognizing database 22, and determines whether the temperature distribution information is within the allowable range. It is determined whether or not it is (S1). When the temperature distribution of the infrared image is within the allowable range, it is determined that the temperature is normal, and the fact is displayed on the display output device 18 (S2). On the other hand, if the device abnormality recognizing means 21 determines that the abnormality is present, an alarm indicating the abnormality is displayed on the display output device (S3).

When an alarm is displayed and output on the display output device 18, the monitor actually reports the abnormality by taking into account whether or not the operation state of the plant equipment has been changed. It is determined whether the status is normal or normal (S4). Even if an alarm is displayed and output, if the monitoring person determines that it is normal, the setting device 2
6 to perform an operation for adding setting of abnormality recognition data (S
5).

As described above, when the operation state of the plant equipment is changed and the temperature distribution is different from the temperature distribution of the abnormality recognition data in the abnormality recognition database 22 but is normal, the abnormality is determined by the judgment of the supervisor. Normal / abnormal of the recognition data is additionally input so that it can be determined that the data is normal.

Since the abnormality recognition database 22 is updated by this additional input, the normal state of the new plant equipment increases. That is, by providing such a learning function, the abnormality recognition database 22 corresponding to a plurality of operating states can be constructed.

[0059]

As described above, according to the present invention, since the infrared image and the visible image of the inspection object are used on the same optical axis, the equipment to be inspected is accurately grasped and its surface temperature distribution is determined. Can be monitored. In addition, since temperature detection can be performed without depending on the surface state of the target device, it is possible to accurately detect abnormalities in the target device with high accuracy regardless of the surface state or operation state of the target device.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a site inspection device according to a first embodiment of the present invention.

FIG. 2 is an explanatory diagram of an image recognition process by an image recognition unit according to the first embodiment of the present invention.

FIG. 3 is an explanatory diagram of device recognition data in a device recognition database according to the first embodiment of the present invention.

FIG. 4 is an explanatory diagram of image processing by an image processing unit according to the first embodiment of the present invention.

FIG. 5 is a configuration diagram of a site inspection device according to a second embodiment of the present invention.

FIG. 6 is a perspective view when the imaging device and the inspection processing device according to the first and second embodiments of the present invention are mounted on a moving device.

FIG. 7 is an explanatory diagram of a site inspection device according to a third embodiment of the present invention.

FIG. 8 is a flowchart showing an operation of the on-site inspection device according to the third embodiment of the present invention when changing data for abnormality recognition.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 inspection target equipment 2 TV camera 3 infrared camera 4 inspection processing device 5 A / D converter 6 inspection processing means 7 image database 8 monitoring device 9 optical device 10 imaging device 11 semi-transparent mirror 12 total reflection mirror 13 digital camera 14 infrared Camera 15 Image recognition means 16 Comparison operation means 17 Device recognition database 18 Display output device 19 Image processing means 20 A / D converter 21 Device abnormality recognition means 22 Abnormality recognition database 23 Moving device 24 Elevation mechanism 25 Control device 26 Setting device

Continued on the front page (51) Int.Cl. ⁷ Identification FI FI Theme Court II (Reference) G21C 17/003 G21C 17/00 E (72) Inventor Masahiko Sasaki 8 Shinsugitacho, Isogo-ku, Yokohama-shi, Kanagawa Prefecture Toshiba Corporation Inside the Yokohama Office (72) Inventor Yoshino Ito 8 Shinsugita-cho, Isogo-ku, Yokohama-shi, Kanagawa Prefecture F-term in the Toshiba Yokohama Office (Reference) 2G051 AA90 AB20 BA06 CA03 CA04 CA07 EA08 EA12 EA14 ED07 ED23 FA10 2G066 AA01 AC07 BA14 BC01 BC07 BC21 CA01 2G075 AA01 BA03 CA02 DA03 EA01 FA13 FB07 FB10 FB12 FB16 FB17 FC04 FC14 GA21 GA34 3J071 CC11 DD36 EE08 EE18 EE27 FF06 FF16 5C084 AA01 AA06 CC19 DD12 DD13 DD61 DD84 EE05 GG17 GG17 GG17 GG17 GG17 GG17 GG17 GG17 GG17 GG17 GG80

Claims (7)

[Claims] 1. An imaging device having a visible wavelength band imaging device for photographing an inspection object and an infrared wavelength band imaging device for photographing a temperature distribution on a surface of the inspection object; A semi-transmissive mirror that transmits the light of the portion and enters the visible wavelength band imaging device and reflects the remaining light, and totally reflects the light reflected by the semi-transmissive mirror and inputs the light to the infrared wavelength band imaging device. An optical device having a total reflection mirror to perform; based on video data of the inspection target device obtained by the visible wavelength band imaging device and video data of the inspection target device obtained by the infrared wavelength band imaging device, An on-site inspection device comprising an inspection processing device for inspecting an abnormality of an inspection object. 2. The image processing device according to claim 1, wherein the inspection processing device extracts a circle component and a line component of the inspection target device shape based on a video signal from the visible wavelength band imaging device, and the image recognition device extracts the circle component and the line component. Comparison operation means for comparing and collating the obtained circle component and line component with the device recognition data in which the characteristics of the device to be inspected are previously described by using the circle component and the line component, and identifying the device to be inspected, and the infrared wavelength band Image processing means for converting into an image having the number of pixels in the direction from the center of gravity as a feature amount after division processing into regions for each temperature based on a video signal from the imaging device, and a feature amount obtained by the image processing means and a normal state The on-site inspection device according to claim 1, further comprising: a device abnormality recognizing unit configured to compare and match a characteristic amount of the temperature distribution of the inspection target device. 3. A visible wavelength band imaging device for photographing an inspection object, and photographing a temperature distribution on a surface of the inspection object.
An imaging device having a first infrared wavelength band imaging device and a second infrared wavelength band imaging device; transmitting a part of light from the inspection object to input to the visible wavelength band imaging device; A first semi-transmissive mirror that reflects the remaining light, and a first infrared wavelength band imaging device that reflects a part of the light reflected by the first semi-transmissive mirror and passes through a first filter And a second filter having a transmission wavelength different from that of the first filter that totally reflects light transmitted through the second semi-transmissive mirror and transmits the remaining light. An optical device having a total reflection mirror for inputting to the second infrared wavelength band imaging device through the camera; video data of the device to be inspected obtained by the visible wavelength band imaging device; and the first infrared wavelength band Imaging device and second infrared wavelength band Field inspection apparatus characterized by comprising an inspection apparatus for inspecting the abnormality of the inspection object based on the image data of the inspection target device obtained by the imaging device. 4. An inspection processing device, comprising: an image recognition unit that extracts a circular component and a line component of the inspection target device shape based on a video signal from the visible wavelength band imaging device; Comparing and comparing the obtained circle component and line component with the device recognition data in which the characteristics of the inspection target device are described in advance by using the circle component and the line component, and specifying the inspection target device; Based on the video signal from the infrared wavelength band imaging device and the video signal from the second infrared wavelength band imaging device, calculate a temperature that does not include the influence of the emissivity of the inspection target device, and divide the temperature into an area for each temperature. Image processing means for converting the number of pixels in the direction from the center of gravity to a feature amount after processing, and comparing and comparing the feature amount obtained by the image processing means with the feature amount of the temperature distribution of the inspection target device in a normal state. Do The on-site inspection device according to claim 3, further comprising a device abnormality recognition unit. 5. The on-site inspection device according to claim 1, further comprising a display output device for displaying and recording an inspection result obtained by the inspection processing device. 6. A moving device that mounts and moves the imaging device, the optical device, and the inspection processing device, and a control device that controls a turning angle and an elevation angle of the imaging device. The on-site inspection device according to claim 1 or 3. 7. A setting device for setting a characteristic quantity of a temperature distribution of the inspection target device in a normal state, which is used for abnormality determination by the device abnormality recognition means, is provided. The on-site inspection device according to claim 4.