Diagnosis device, method for diagnosis, and program

Abstract

To provide a diagnosis device for properly operating the surface temperature of a target object.SOLUTION: A diagnosis device is for diagnosing a target object, and includes a storage unit for storing a thermal image obtained by taking an image of an object by a thermal image camera and the radiation rate of the target range of the target object and a calculation unit for calculating the surface temperature of the target range on the basis of the thermal image and the radiation rate.SELECTED DRAWING: Figure 4

Description

Technical Field

[0001] This specification relates to a diagnostic apparatus, a diagnostic method, and a program.

Background Art

[0002] Conventionally, for example, a diagnostic apparatus calculates the surface temperature of an object based on a thermal image captured by a thermal imaging camera of the object such as a pipe (for example, Patent Document 1). By the way, even if the actual surface temperature of the object is the same, since the infrared energy radiated from the object varies depending on the emissivity of the object, the calculated surface temperature of the object may be different from the actual surface temperature of the object.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Therefore, the problem is to provide a diagnostic apparatus, a diagnostic method, and a program capable of appropriately calculating the surface temperature of an object.

Means for Solving the Problems

[0005] [1] The diagnostic apparatus is a diagnostic apparatus for diagnosing an object, a storage unit that stores a thermal image of the object captured by a thermal imaging camera and the emissivity of the target range of the object; and a calculation unit that calculates the surface temperature of the target range based on the thermal image and the emissivity.

[0006] [2] In the diagnostic apparatus of [1] above, The memory unit stores the measured temperature obtained by measuring the surface of the target position within the target range with a thermometer, The calculation unit calculates the emissivity of the target range based on the thermal image and the measured temperature, The memory unit stores the calculated emissivity, Such a configuration may be used.

[0007] [3] In the diagnostic apparatus of the above [2], The calculation unit calculates the target range so as to include the target position based on the thermal image, Such a configuration may be used.

[0008] [4] In the diagnostic apparatus of the above [3], The calculation unit Uses the target range setting thermal image, which is the thermal image with the target range set, as teacher data, and machine-learns a determination model that takes the thermal image as input and outputs the target range calculation thermal image, which is the thermal image with the target range calculated, Outputs the target range calculation thermal image from a specific thermal image using the determination model, Such a configuration may be used.

[0009] [5] In any one of the diagnostic apparatuses of the above [2] to [4], The memory unit stores the visible light image obtained by photographing the object with a visible light camera, The calculation unit calculates the target range so as to include the target position based on the thermal image and the visible light image, Such a configuration may be used.

[0010] [6] In any one of the diagnostic apparatuses of the above [3] to [5], The calculation unit Based on the thermal image, divides the target range into a first small range including the target position and a second small range determined to have an emissivity different from that of the first small range, Based on the thermal image and the measured temperature, calculate the emissivity of the first small range. Based on the thermal image and the emissivity of the first small range, calculate the surface temperature of the first small range. Such a configuration may be used.

[0011] [7] In the diagnostic apparatus of the above [3] or [4], The storage unit stores a visible light image obtained by photographing the object with a visible light camera. The calculation unit Based on the thermal image and the visible light image, divide the target range into a first small range including the target position and a second small range determined to have an emissivity different from that of the first small range. Based on the thermal image and the measured temperature, calculate the emissivity of the first small range. Based on the thermal image and the emissivity of the first small range, calculate the surface temperature of the first small range. Such a configuration may be used.

[0012] [8] In the diagnostic apparatus of the above [5], The calculation unit Based on the thermal image and the visible light image, divide the target range into a first small range including the target position and a second small range determined to have an emissivity different from that of the first small range. Based on the thermal image and the measured temperature, calculate the emissivity of the first small range. Based on the thermal image and the emissivity of the first small range, calculate the surface temperature of the first small range. Such a configuration may be used.

[0013] [9] In any one of the diagnostic apparatuses of the above [6] to [8], The calculation unit calculates the surface temperature of the second small range based on the surface temperature of the first small range. Such a configuration may be used.

[0014]

[10] Any one of the diagnostic devices [1] to [9] described above is further provided with an input unit to which data is input, wherein the input unit receives the emissivity of the target range, and the storage unit stores the input emissivity. Such a configuration may be used.

[0015]

[11] In any one of the diagnostic devices [1] to

[10] described above, the storage unit stores a visible light image obtained by photographing the object with a visible light camera and a measured temperature obtained by measuring the surface of the target position within the target range with a thermometer, the diagnostic device includes a display unit for displaying the thermal image and the visible light image, and an input unit for inputting the position of the target position with respect to the visible light image displayed on the display unit, wherein the calculation unit calculates the position of the thermal image corresponding to the position of the target position input to the input unit based on the thermal image and the visible light image. Such a configuration may be used.

[0016]

[12] In the diagnostic device of

[11] described above, the display unit displays the position calculated by the calculation unit on the thermal image. Such a configuration may be used.

[0017]

[13] A diagnostic method is a diagnostic method for diagnosing an object using any one of the diagnostic devices [1] to

[12] described above, including storing a thermal image obtained by photographing the object with a thermal imaging camera, storing the emissivity of the target range of the object, and calculating the surface temperature of the target range based on the thermal image and the emissivity.

[0018]

[14] The program causes at least one processor to execute the diagnostic method of

[13] above.

Brief Description of the Drawings

[0019]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Modes for Carrying Out the Invention

[0020] In each drawing, for example, for ease of understanding, the dimensions of the components may be shown enlarged or reduced with respect to the actual dimensions, and the dimensional ratios between the drawings may not match. In each drawing, for example, for ease of understanding, a part of the components may be shown omitted.

[0021] Terms including ordinal numbers such as first, second, etc. are used to describe various components, but this term is used only for the purpose of distinguishing one component from another, and the components are not particularly limited by this term. Note that the number of components including ordinal numbers is not particularly limited, and for example, there may be one. Also, the ordinal numbers used in the following specification and drawings may be different from those described in the claims.

[0022] Hereinafter, an embodiment of the diagnostic system and the diagnostic device will be described with reference to FIGS. 1 to 11. Note that the following embodiments are exemplified to assist in understanding the configuration and the like of the diagnostic system and the diagnostic device, and do not limit the configuration of the diagnostic system and the diagnostic device.

[0023] As shown in FIG. 1, the diagnostic system 10 may include, for example, a diagnostic device 1 for diagnosing an object (for example, a pipe), a portable terminal device 2, a thermometer 3 for measuring the temperature of the object, and a storage device 4 for storing data. The diagnostic device 1, the terminal device 2, and the storage device 4 may be communicable with each other by communication means 5, for example, as in this embodiment.

[0024] Although not particularly limited, the communication means 5 may be, for example, wired communication means such as a wired LAN or a communication cable, or may be wireless communication means such as the Internet, Bluetooth, or a wireless LAN. Note that, for example, the thermometer 3 may also be communicable with each of the devices 1, 2, and 4 by the communication means 5.

[0025] The memory device 4 may be, for example, a server such as a cloud server. Also, although not particularly limited, the owner of the memory device 4 may be the same as, for example, the owners of the diagnostic device 1, the terminal device 2, and the thermometer 3, or may be different.

[0026] The thermometer 3 is not particularly limited as long as it can measure the surface temperature of a specific position (spot) of the object. The thermometer 3 is preferably, for example, a contact-type thermometer 3 that can measure the temperature in contact with the surface of the object.

[0027] As shown in FIGS. 2 and 3, the terminal device 2 may include, for example, a visible light camera 6 and a thermal image camera 7 that capture an object, an input unit (also referred to as a "terminal input unit") 21 to which data is input, a processing unit (also referred to as a "terminal processing unit") 22 that processes data, and an output unit (also referred to as a "terminal output unit") 23 that outputs data.

[0028] The terminal processing unit 22 may include, for example, an acquisition unit (also referred to as a "terminal acquisition unit") 24 that acquires data, a storage unit (also referred to as a "terminal storage unit") 25 that stores data, an arithmetic unit (referred to as a "terminal arithmetic unit") 26 that performs arithmetic operations on the data, and a control unit (also referred to as a "terminal control unit") 27 that controls each part of the terminal device 2.

[0029] Also, for example, as in the present embodiment, the terminal input unit 21 includes a touch panel 21a and a button 21b, and the terminal output unit 23 may be configured to include a display unit (display) 23a that displays data and a sound output unit 23b that emits data as sound. The touch panel 21a is formed transparently and is disposed on the surface of the display unit 23a, and the display of the display unit 23a is visible through the touch panel 21a.

[0030] Thus, the terminal device 2 may be a tablet computer as in this embodiment. Note that the terminal device 2 is not limited to such a configuration, and may be, for example, a notebook personal computer, or may be, for example, a smart device (smartphone).

[0031] The visible light camera 6 receives, for example, visible light radiated from and reflected by an object and detects it, and generates a visible light image based on the detected visible light. Although not particularly limited, the visible light camera 6 may be, for example, a CCD camera or a CMOS camera.

[0032] The thermal image camera 7 detects, for example, infrared energy radiated from and reflected by an object, converts the detected infrared energy into temperature, and generates a thermal image visualized as an image representing a temperature distribution. Although not particularly limited, the thermal image camera 7 may be, for example, a thermographic camera.

[0033] In this embodiment, the visible light camera 6 and the thermal image camera 7 are configured to be able to photograph an object at the same timing. Note that although not particularly limited, for example, as in this embodiment, the visible light camera 6 may be non-removably attached to the main body 2a of the terminal device 2, and the thermal image camera 7 may be removably attached to the main body 2a of the terminal device 2 and connected to the main body 2a of the terminal device 2 via a cable 2b.

[0034] The terminal input unit 21 may include, for example, as in this embodiment, a shooting instruction input unit 21c to which data for instructing shooting (shooting instruction data) is input, a measured temperature input unit 21d to which data of the temperature measured by the thermometer 3 (measured temperature data) is input, and a measured position input unit 21e to which data of the position measured by the thermometer 3 (measured position data) is input.

[0035] As shown in FIG. 4, the diagnostic device 1 may include, for example, an input unit (also referred to as "device input unit") 11 to which data is input, a processing unit (also referred to as "device processing unit") 12 that processes data, and an output unit (also referred to as "device output unit") 13 that outputs data.

[0036] The device processing unit 12 may include, for example, an acquisition unit (also referred to as "device acquisition unit") 14 that acquires data, a storage unit (also referred to as "device storage unit") 15 that stores data, an arithmetic unit (also referred to as "device arithmetic unit") 16 that performs arithmetic operations on the data, and a control unit (also referred to as "device control unit") 17 that controls each part of the diagnostic device 1, as in this embodiment.

[0037] Although not particularly limited, the diagnostic device 1 may be, for example, a desktop personal computer installed at a predetermined position. Also, the diagnostic device 1 may be, for example, a notebook personal computer, a tablet computer, or a smart device (smartphone) that can be carried and transported.

[0038] Also, the diagnostic device 1 may be composed of, for example, a single device, or may be composed of, for example, a plurality of devices capable of communicating with each other. Specifically, each part 11, 12, 13 of the diagnostic device 1 may be provided in, for example, a single device, or may be distributed and provided in a plurality of devices capable of communicating with each other.

[0039] The configuration of the device input unit 11 is not particularly limited, but the device input unit 11 may be, for example, a mouse, a keyboard, a touch panel, a switch (e.g., a push button switch, a select switch, etc.).

[0040] The configuration of the device output unit 13 is not particularly limited, but the device output unit 13 may include, for example, a display unit 13a that displays data (e.g., a monitor, an electronic bulletin board, an indicator light), a sound output unit 13b that emits data as sound (e.g., a buzzer, a speaker, etc.), and a transmission unit (not shown) that transmits data to the outside of the diagnostic device 1.

[0041] The device processing unit 12 may be, for example, a computer including processors such as a CPU and an MPU (for example, the device arithmetic unit 16 and the device control unit 17), memories such as a ROM and a RAM (for example, the device acquisition unit 14 and the device storage unit 15), various interfaces, and the like. Thereby, the processor executes the program stored in the memory, and each of the units 14, 15, 16, and 17 of the device processing unit 12 is realized by the cooperation of software and hardware.

[0042] Note that the device processing unit 12 may be configured by, for example, a single device, or may be configured by, for example, a plurality of devices capable of communicating with each other. Specifically, each of the units 14, 15, 16, and 17 of the device processing unit 12 may be provided in, for example, a single device, or may be distributed and provided in, for example, a plurality of devices capable of communicating with each other.

[0043] The device storage unit 15 may include, for example, an image storage unit 15a, a temperature storage unit 15b, an emissivity storage unit 15c, and a model storage unit 15d as in the present embodiment. The device arithmetic unit 16 may include, for example, a target range arithmetic unit 16a, an emissivity arithmetic unit 16b, a surface temperature arithmetic unit 16c, a heat retention state diagnosis unit 16d, and a learning unit 16e as in the present embodiment.

[0044] Note that in this specification, storage means not only storing continuously until deleted by the input of a deletion instruction, but also temporarily storing for use in the operations of the arithmetic units 16 and 26 (for example, acquiring from the storage device 4 (see FIG. 1) and temporarily storing), and including the case of being deleted after the operation ends regardless of the input of a deletion instruction.

[0045] Next, a diagnosis method using the diagnosis system 10 and the diagnosis device 1 according to the present embodiment will be described with reference to FIGS. 5 to 8. Note that the following method is an example for helping the understanding of the diagnosis method and the like using the diagnosis system 10 and the diagnosis device 1, and does not limit the diagnosis method using the diagnosis system 10 and the diagnosis device 1.

[0046] For example, as one of the objects to be diagnosed by the diagnostic device 1, as shown in FIG. 5, there is a pipe 9. The pipe 9 may include, for example, a pipe material 91 through which a fluid flows inside, a heat insulating material 92 covering the pipe material 91, and a cover material 93 covering the heat insulating material 92. In the case of the pipe 9 having the heat insulating material 92, for example, the temperature of the fluid may be significantly higher or lower than the ambient temperature of the pipe 9.

[0047] Examples of the fluid having a temperature significantly higher than the ambient temperature of the pipe 9 include, for example, steam exceeding 100°C, hot water at 70°C to 90°C, etc. Examples of the fluid having a temperature significantly lower than the ambient temperature of the pipe 9 include, for example, cold water at 5°C to 10°C, chiller water below 0°C, etc. Although not particularly limited, in this embodiment, the object to be diagnosed is the pipe 9 through which steam flows inside.

[0048] As shown in FIG. 6, in the photographing step S1, for example, when photographing instruction data is input to the photographing instruction input unit 21c, each of the visible light camera 6 and the thermal image camera 7 photographs the pipe 9. Thereby, for example, as shown in FIG. 7, the display unit 23a of the terminal device 2 displays a visible light image (the left image in FIG. 7) of the pipe 9 photographed by the visible light camera 6 and a thermal image (the right image in FIG. 7) of the pipe 9 photographed by the thermal image camera 7.

[0049] Note that the device arithmetic unit 16 may be configured to process the visible light image and the thermal image (for example, trim, enlarge, or reduce) so that the sizes of the objects in both images are the same. Also, although not particularly limited, each of the input units 11, 21, 21c to 21e may be configured such that data can be input, for example, by touching a predetermined position of the display units 13a, 23a, or, for example, by aligning the mouse pointer with a predetermined position of the display units 13a, 23a and clicking the mouse to input data.

[0050] Next, in the temperature measurement step S2, for example, the surface temperature of the target position of the pipe 9 is measured by the thermometer 3. Then, for example, the measured temperature data is input to the measured temperature input unit 21d, and the measured position data is input to the measured position input unit 21e. As a result, for example, as shown in FIG. 7, the display unit 23a of the terminal device 2 displays the surface temperature of the target position of the pipe 9 (the temperature displayed as "measured temperature" in FIG. 7) and the position of the target position (the position of the "+", i.e., the center of the crosshair, in the thermal image in FIG. 7).

[0051] Although not particularly limited, for example, the measured position input unit 21e may be configured such that the position of the target position is input with respect to the thermal image displayed on the display unit 23a. Also, for example, the measured position input unit 21e may be configured such that the position of the target position is input with respect to the visible light image displayed on the display unit 23a, and the terminal arithmetic unit 26 calculates the position of the thermal image corresponding to the position of the target position input to the measured position input unit 21e.

[0052] Then, the terminal storage unit 25 stores the visible light image and the thermal image captured in the imaging step S1 in association with the measured temperature of the target position and the position of the target position input in the temperature measurement step S2. Note that, if necessary, the imaging step S1 and the temperature measurement step S2 may be repeated.

[0053] Thereafter, the terminal output unit 23 outputs the data stored in association to the diagnostic device 1. As a result, the device storage unit 15 stores the data output from the terminal device 2. For example, the image storage unit 15a stores the visible light image and the thermal image, and the temperature storage unit 15b stores the surface temperature of the target position and the position of the target position.

[0054] Next, in the target range calculation step S3, for example, as shown in FIG. 8, the target range calculation unit 16a calculates a target range A0 so as to include the target position based on the thermal image. As a result, the target range A0 can be set. Then, for example, the thermal image for which the target range A0 has been calculated is displayed on the display unit 13a of the device output unit 13.

[0055] In Fig. 8, the target position is the position of "+", and the target range A0 is the range surrounded by the solid line. Also, for example, when the calculated target range A0 is not appropriate, the target range calculation unit 16a may correct the target range A0 by inputting correction instruction data to the device input unit 11.

[0056] Although not particularly limited, examples of the method for calculating the target range A0 by the target range calculation unit 16a include, for example, the following two examples.

[0057] As a first example, the target range calculation unit 16a may be configured such that, for example, the range where the temperature of the thermal image is equal to or higher than the threshold value is set as the target range A0. Thereby, the target range calculation unit 16a calculates the target range A0 based on the thermal image P2a. Although not particularly limited, the threshold value may be, for example, the ambient temperature (e.g., air temperature), or the temperature obtained by adding the set temperature (e.g., 1°C to 5°C) to the ambient temperature. Also, although not particularly limited, the ambient temperature may be input to the terminal input unit 21 and stored in the device storage unit 15 (temperature storage unit 15b).

[0058] Also, as a second example, a calculation method using machine learning may be used. For example, as shown in Fig. 9, the learning unit 16e uses the target range setting image P1 with the target range A0 set as the teacher data, inputs the image P2, and outputs the target range calculation image P3 which is an image in which the target range A0 is calculated. The learning unit 16e performs machine learning on the target range determination model, and the target range calculation unit 16a may be configured to output the target range calculation image P3 from a specific image P2 using the target range determination model.

[0059] Although not particularly limited, for example, the learning unit 16e may learn a neural network by deep learning. Also, although not particularly limited, for example, the target range calculation unit 16a may output the target range calculation image P3 from a specific image P2 using the learned neural network. The model storage unit 15d may store, for example, the target range determination model machine-learned by the learning unit 16e.

[0060] Note that the combination of the target range setting image P1, the image P2, and the target range calculation image P3 is not particularly limited, and examples include the following first to third combinations and the like.

[0061] As the first combination, for example, the target range setting image P1, which is teacher data, may be a thermal image (target range setting thermal image) P1a in which the target range A0 is set, the input image P2 may be a thermal image P2a, and the output target range calculation image P3 may be a thermal image (target range calculation thermal image) P3a in which the target range A0 is calculated.

[0062] In such a configuration, the target range calculation unit 16a outputs the target range calculation thermal image P3a from the thermal image P2a using the target range determination model. Thereby, the target range calculation unit 16a calculates the target range A0 based on the thermal image P2a.

[0063] Note that when the object is the pipe 9 for a heating medium (for example, steam), in the target range setting thermal image P1a, which is teacher data, for example, there are features such as the target range A0 being higher in temperature than the surroundings, a large temperature difference at the boundary between the target range A0 and the other range, and the shape of the target range A0 being composed of a straight portion (straight pipe) or a bent portion (elbow).

[0064] Also, as the second combination, for example, the target range setting image P1, which is teacher data, may be a visible light image (target range setting visible light image) P1b in which the target range A0 is set, the input image P2 may be a visible light image P2b, and the output target range calculation image P3 may be a visible light image (target range calculation visible light image) P3b in which the target range A0 is calculated.

[0065] In such a configuration, the target range calculation unit 16a outputs a target range calculation visible light image P3b from the visible light image P2b using the target range determination model. Then, the target range calculation unit 16a calculates a target range calculation thermal image P3a based on the thermal image 2a (capturing the pipe 9 in the same area) associated with the visible light image P2b and the target range calculation visible light image P3b. Therefore, the target range calculation unit 16a calculates the target range A0 based on the associated visible light image P2b and thermal image P2a.

[0066] In addition, when the object is the insulated pipe 9, in the target range setting visible light image P1b which is the teacher data, for example, there are features such as the color of the target range A0 being a specific color (for example, the color of the covering material 93, which is silver), the amount of change in the color at the boundary between the target range A0 and the other range being large, and the shape of the target range A0 being composed of a straight part (straight pipe) and a bent part (elbow).

[0067] Also, as a third combination, for example, the target range setting image P1 which is the teacher data may be the associated target range setting thermal image P1a and target range setting visible light image P1b (capturing the pipe 9 in the same area), the input image P2 may be the associated visible light image P2b and thermal image P2a (capturing the pipe 9 in the same area), and the output target range calculation image P3 may be the target range calculation thermal image P3a and target range calculation visible light image P3b.

[0068] In such a configuration, the target range calculation unit 16a outputs a target range calculation thermal image P3a and a target range calculation visible light image P3b from the associated thermal image P2a and visible light image P2b using, for example, the target range determination model. Thereby, the target range calculation unit 16a calculates the target range A0 based on the associated visible light image P2b and thermal image P2a.

[0069] In the associated (piping 9 in the same area photographed) target range setting thermal image P1a and target range setting visible light image P1b, which are teacher data, for example, in addition to the above-described features, there are features such as the target ranges A0 set for each other being the same range (that is, the target range A0 set in the target range setting thermal image P1a and the target range A0 set in the target range setting visible light image P1b are the same range).

[0070] Thus, for example, according to the calculation methods such as the above two examples, the calculation unit 16 can calculate the target range A0 so as to include the target position based on the thermal image 2a or based on the thermal image 2a and the visible light image 2b.

[0071] Next, in the small range calculation step S4, the target range calculation unit 16a divides the target range A0 into, for example, as shown in FIG. 10, a first small range A1 including the target position and a second small range A2 determined to have an emissivity different from that of the first small range A1. Thereby, the first small range A1 and the second small range A2 having different emissivities can be set. Then, for example, the thermal images in which the small ranges A1 and A2 are calculated are displayed on the display unit 13a of the device output unit 13.

[0072] For example, even in the same piping system, due to differences in the covering material 93 or partial painting of the covering material 93, the emissivity is different, so such a range is set as the second small range A2. In FIG. 10, the boundary line between the first small range A1 and the second small range A2 is a broken line.

[0073] In FIG. 10, there is only one second small range A2, but for example, there may be two or more, or for example, there may be none. Also, for example, if the divided first small range A1 and second small range A2 are not appropriate, correction instruction data is input to the device input unit 11, and the target range calculation unit 16a may correct the division of the first small range A1 and the second small range A2.

[0074] Although not particularly limited, as an example of the calculation method for dividing the first small range A1 and the second small range A2 by the target range calculation unit 16a, for example, the following four examples can be cited.

[0075] As a first example, the target range calculation unit 16a may be configured to, for example, determine that the emissivity is different at the boundary when the temperature difference at the boundary in the thermal image is equal to or greater than a threshold value, and divide it into the first small range A1 and the second small range A2. Thereby, the target range calculation unit 16a divides the first small range A1 and the second small range A2 based on the thermal image.

[0076] Also, as a second example, the target range calculation unit 16a may be configured to, for example, determine that the emissivity is different at the boundary when the amount of change in the color at the boundary in the visible light image is equal to or greater than a threshold value, and divide it into the first small range A1 and the second small range A2. Thereby, the target range calculation unit 16a divides the first small range A1 and the second small range A2 based on the visible light image.

[0077] Also, as a third example, the target range calculation unit 16a may be configured to, for example, determine that the emissivity is different at the boundary when the temperature difference at the boundary in the thermal image is equal to or greater than a threshold value and the amount of change in the color at the boundary in the visible light image is equal to or greater than a threshold value, and divide it into the first small range A1 and the second small range A2. Thereby, the target range calculation unit 16a divides the first small range A1 and the second small range A2 based on the thermal image and the visible light image.

[0078] Also, as a fourth example, a calculation method using machine learning may be used. For example, as shown in FIG. 11, the learning unit 16e uses the small range setting image P4 in which the small ranges A1 and A2 are set as teacher data, inputs the image P5, and outputs the small range calculation thermal image P6 which is an image in which the small ranges A1 and A2 are calculated, and machine-learns a small range determination model. The target range calculation unit 16a may be configured to output the small range calculation image P6 from a specific image P5 using the small range determination model.

[0079] Although not particularly limited, for example, the target range calculation unit 16a may output a small range calculation image P6 from a specific image P5 using a learned neural network. Note that the model storage unit 15d may store, for example, a small range determination model learned by the learning unit 16e.

[0080] Note that the combination of the small range setting image P4, the image P5, and the small range calculation image P6 is not particularly limited, and examples include the following first to third combinations.

[0081] As a first combination, for example, the small range setting image P4, which is teacher data, is a thermal image (small range setting thermal image) P4a in which small ranges A1 and A2 are set, the input image P5 is a thermal image P5a (specifically, a target range calculation thermal image P3a), and the output small range calculation image P6 may have a configuration in which a thermal image (small range calculation thermal image) P6a in which the small ranges A1 and A2 are calculated.

[0082] In such a configuration, the target range calculation unit 16a outputs a small range calculation thermal image P6a from the target range calculation thermal image P3a using the small range determination model. As a result, the target range calculation unit 16a divides the small ranges A1 and A2 based on the thermal image P5a (for example, the target range calculation thermal image P3a).

[0083] Note that when the object is the insulated pipe 9, in the small range setting thermal image P4a, which is teacher data, for example, there are features such as a large temperature difference at the boundary between the first small range A1 and the second small range A2, or when the cover material 93 is different, the boundary between the first small range A1 and the second small range A2 is perpendicular to the direction in which the pipe 9 extends.

[0084] Also, as a second combination, for example, the small-range setting image P4, which is teacher data, is a visible light image (small-range setting visible light image) P4b in which small ranges A1 and A2 are set, the input image P5 is a visible light image 5b (specifically, the target range calculation visible light image P3b), and the output small-range calculation image P6 may be a visible light image (small-range calculation visible light image) P6b in which the small ranges A1 and A2 are calculated.

[0085] In such a configuration, the target range calculation unit 16a outputs a small-range calculation visible light image P6b from the target range calculation visible light image P3b using a small-range determination model. Then, the target range calculation unit 16a calculates a target range calculation thermal image P6a based on the target range calculation thermal image 3a (capturing the pipe 9 in the same area) associated with the target range calculation visible light image P3b and the small-range calculation visible light image P6b. Therefore, the target range calculation unit 16a divides the small ranges A1 and A2 based on the associated thermal image P5a and visible light image P5b (for example, the target range calculation thermal image P3a and the target range calculation visible light image P3b).

[0086] When the object is the insulated pipe 9, in the small-range setting visible light image P4b, which is teacher data, for example, there are features such as a large change amount in the color at the boundary between the first small range A1 and the second small range A2, or when the cover material 93 is different, the boundary between the first small range A1 and the second small range A2 is perpendicular to the extending direction of the pipe 9.

[0087] Also, as a third combination, for example, the small-range setting image P4, which is teacher data, is a small-range setting thermal image P4a and a small-range setting visible light image P4b associated with each other (capturing the pipe 9 in the same area), the input image P5 is a thermal image P5a and a visible light image P5b associated with each other (capturing the pipe 9 in the same area) (specifically, the target range calculation thermal image P3a and the target range calculation visible light image P3b), and the output target range calculation image P6 may be a small-range calculation thermal image P6a and a small-range calculation visible light image P6b.

[0088] In such a configuration, the target range calculation unit 16a outputs, for example, a small-range calculated thermal image P6a and a small-range calculated visible light image P6b from the associated target range calculated thermal image P3a and target range calculated visible light image P3b using, for example, a small-range determination model. Thereby, the target range calculation unit 16a divides the small ranges A1 and A2 based on the associated thermal image P5a and visible light image P5b (for example, the target range calculated thermal image P3a and target range calculated visible light image P3b).

[0089] In the case of the associated (photographing the same area of the pipe 9) small-range set thermal image P4a and small-range set visible light image P4b which are teacher data, for example, in addition to the above-described features, there are features such that the small ranges A1 and A2 set for each other are in the same range (that is, the small ranges A1 and A2 set in the small-range set thermal image P4a and the small ranges A1 and A2 set in the small-range set visible light image P4b are in the same range respectively).

[0090] Thus, for example, according to the calculation method such as the above four examples, the calculation unit 16 can divide the target range A0 into a first small range A1 including the target position and a second small range A2 determined to have an emissivity different from that of the first small range A1 based on the thermal image P5a (for example, the target range calculated thermal image P3a), or based on the thermal image P5a and the visible light image P5b (for example, the target range calculated thermal image P3a and target range calculated visible light image P3b).

[0091] Next, in the emissivity calculation step S5, the emissivity calculation unit 16b calculates, for example, the emissivity of the first small range A1 based on the thermal image and the surface temperature of the target position, and the emissivity storage unit 15c stores the calculated emissivity. Thereby, the emissivity of the first small range A1 can be appropriately calculated.

[0092] Although not particularly limited, the emissivity calculation unit 16b may be configured to calculate the emissivity of the target position based on, for example, the temperature of the target position in the thermal image, the measured surface temperature of the target position, and the Stefan-Boltzmann law. Specifically, the emissivity calculation unit 16b may be configured to calculate the emissivity of the target position by, for example, the following Equation 1.

[0093] (T1 + 273.15) × (emissivity of the target position) -0.25 −273.15 = T2 (Equation 1) However, T1 (assumed temperature) is the temperature (°C) of the target position in the thermal image when the emissivity of the target position is assumed to be 1.0, that is, when the target object is assumed to be a blackbody, and T2 (measured temperature) is the measured surface temperature (°C) of the target position.

[0094] By the way, the thermal image camera 7 captures a thermal image with the emissivity of the object to be captured set to a predetermined emissivity (hereinafter also referred to as "camera-set emissivity"). Therefore, when the camera-set emissivity is not 1.0, the emissivity calculation unit 16b may be configured to calculate T1 (assumed temperature) of Equation 1 above by, for example, the following Equation 2. (T3 + 273.15) × (camera-set emissivity) -0.25 −273.15 = T1 (Equation 2) However, T3 (assumed temperature) is the temperature (°C) of the target position in the thermal image when the emissivity of the target position is assumed to be the camera-set emissivity.

[0095] Note that the emissivity storage unit 15c may store, for example, Equation 1 and Equation 2 above. Further, the emissivity storage unit 15c may store, for example, the emissivity of the calculated first small range A1. Also, the set emissivity of the thermal image camera may be input to the input unit 11, and the emissivity storage unit 15c may store, for example, the input set emissivity of the thermal image camera.

[0096] Next, in the surface temperature calculation step S6, the surface temperature calculation unit 16c calculates the surface temperature of the first small range A1 based on the thermal image stored in the image storage unit 15a and the emissivity of the first small range A1 stored in the emissivity storage unit 15c. Thereby, the surface temperature of the first small range A1 can be appropriately calculated. Although not particularly limited, the surface temperature calculation unit 16c may calculate one surface temperature for each pixel of the thermal image, for example, or may calculate one surface temperature by averaging several pixels of the thermal image, for example.

[0097] Also, although not particularly limited, the surface temperature calculation unit 16c may be configured to calculate the surface temperature of the first small range A1 based on, for example, the temperature of the first small range A1 of the thermal image, the emissivity of the first small range A1, and the Stefan-Boltzmann law. Specifically, the surface temperature calculation unit 16c may be configured to calculate the surface temperature (°C) of the first small range A1 by, for example, the following equation 3.

[0098] Surface temperature = (T4 + 273.15) × (emissivity of the first small range A1) -0.25 -273.15 (Equation 3) However, T4 (assumed temperature) is the temperature (°C) of the first small range A1 of the thermal image when the emissivity of the first small range A1 is assumed to be 1.0, that is, when the first small range A1 is assumed to be a blackbody.

[0099] When the camera-set emissivity is not 1.0, the surface temperature calculation unit 16c may be configured to calculate T4 (assumed temperature) of the above equation 3 by, for example, the following equation 4. (T5 + 273.15) × (camera-set emissivity) -0.25 -273.15 = T4 (Equation 4) However, T5 (assumed temperature) is the temperature (°C) of the first small range A1 of the thermal image when the emissivity of the first small range A1 is assumed to be the camera-set emissivity.

[0100] After that, for example, based on the calculated surface temperature of the first small range A1, the surface temperature calculation unit 16c calculates the surface temperature of the second small range A2. Thereby, the surface temperature of the second small range A2 can be appropriately calculated.

[0101] Although not particularly limited, as an example of the method for calculating the surface temperature of the second small range A2 by the surface temperature calculation unit 16c, for example, the following four examples can be cited.

[0102] As a first example, for example, the surface temperature of the second small range A2 is likely to be close to the average of the surface temperatures of the first small range A1. Therefore, the surface temperature calculation unit 16c may be configured such that the temperature of the second small range A2 is the average of the surface temperatures of the adjacent first small ranges A1.

[0103] Also, as a second example, for example, when the surface temperatures of the regions adjacent to the second small range A2 among the first small ranges A1, A1 sandwiching the second small range A2 are the same, the surface temperature of the second small range A2 is likely to be the same as the surface temperatures of the adjacent regions. Therefore, in such a case, the surface temperature calculation unit 16c may be configured such that the surface temperature of the second small range A2 is the same as the surface temperatures of the adjacent first small ranges A1, A1.

[0104] Also, as a third example, for example, the emissivity of the pipe 9 largely depends on the material of the cover material 93. And if the material of the cover material 93 is known, the difference between the design value of the emissivity of the cover material 93 in the first small range A1 and the design value of the emissivity of the cover material 93 in the second small range A2 can be known. Therefore, for example, the difference is input to the input units 11, 21, and the surface temperature calculation unit 16c calculates the emissivity of the second small range A2 based on the calculated emissivity of the first small range A1 and the input difference, and calculates the surface temperature of the second small range A2 based on the thermal image and the calculated emissivity of the second small range A2. Such a configuration may also be used.

[0105] Also, as a fourth example, for instance, the materials of the covering material 93 of the first small range A1 and the second small range A2 are input to the input units 11 and 21. The surface temperature calculation unit 16c calculates the difference in the design value of the emissivity of the covering material 93 from the input materials, calculates the emissivity of the second small range A2 based on the calculated emissivity of the first small range A1 and the calculated difference, and calculates the surface temperature of the second small range A2 based on the thermal image and the calculated emissivity of the second small range A2. Such a configuration may be adopted.

[0106] Thus, for example, according to the calculation methods such as the above four examples, the calculation unit 16 can calculate the surface temperature of the second small range A2 based on the surface temperature of the first small range A1.

[0107] Note that the temperature storage unit 15b may store, for example, the above formulas 3 and 4. Also, the temperature storage unit 15b may store, for example, various data necessary for calculating the surface temperature of the second small range A2. Further, the temperature storage unit 15b may store, for example, the calculated surface temperature, specifically, the respective surface temperatures of the calculated first small range A1 and the second small range A2.

[0108] Next, in the heat insulation state diagnosis step S7, the heat insulation state diagnosis unit 16d diagnoses the heat insulation state of the pipe 9 based on, for example, the surface temperature of the target range A0 calculated in the surface temperature calculation step S5. Thereby, the heat insulation state of the pipe 9 can be appropriately diagnosed. Note that the diagnosis method of the heat insulation state diagnosis unit 16d is not particularly limited.

[0109] For example, the heat insulation state diagnosis unit 16d may diagnose the heat insulation state of the pipe 9 by comparing the calculated surface temperature of the target range A0 with the design value of the surface temperature of the target range A0. Also, for example, the heat insulation state diagnosis unit 16d may calculate the heat dissipation amount from the calculated surface temperature of the target range A0 and diagnose the heat insulation state of the pipe 9 by comparing the calculated heat dissipation amount with the design value of the heat dissipation amount of the target range A0.

[0110] In this way, in the surface temperature calculation step S6, by considering the emissivity of the target range A0 of the object, the surface temperature of the object can be appropriately calculated. Therefore, for example, in the heat insulation state diagnosis step S7, the heat insulation state can be accurately diagnosed. Thus, for example, heat leakage of the object due to inappropriate heat insulation can be diagnosed with high accuracy.

[0111] Note that, while the method of diagnosing the heat insulation state of the pipe 9 has been described with reference to FIGS. 5 to 11, the method is not limited thereto. For example, a pipe having no heat insulating material 92 and cover material 93, that is, the pipe material 91 may be photographed, the surface temperature of the pipe material 91 may be calculated, and the heat dissipation amount from the pipe material 91 may be calculated to diagnose the heat loss situation. Further, for example, a device that generates heat may be photographed, the surface temperature of the device may be calculated, and the heat distribution situation of the device may be diagnosed.

[0112] From the above, the diagnostic device 1 is, as in this embodiment, a diagnostic device 1 for diagnosing an object, a storage unit 15 that stores a thermal image of the object photographed by the thermal imaging camera 7 and the emissivity of the target range A0 of the object; and a calculation unit 16 that calculates the surface temperature of the target range A0 based on the thermal image and the emissivity. Such a configuration is preferable.

[0113] According to such a configuration, based on the thermal image of the object photographed by the thermal imaging camera 7 and the emissivity of the target range A0 of the object, the surface temperature of the target range A0 of the object is calculated. Thereby, since the emissivity of the target range A0 of the object is considered, the surface temperature of the object can be appropriately calculated.

[0114] Also, in the diagnostic device 1, as in this embodiment, the storage unit 15 stores the measured temperature measured by the thermometer 3 on the surface of the target position within the target range A0, and the calculation unit 16 calculates the emissivity of the target range A0 based on the thermal image and the measured temperature. The memory unit 15 stores the calculated emissivity. Such a configuration is preferable.

[0115] According to such a configuration, based on the thermal image and the measured temperature measured by the thermometer 3 on the surface of the target position within the target range A0, the emissivity of the target range A0 is calculated. Thereby, the emissivity of the target range A0 can be appropriately calculated.

[0116] Also, in the diagnostic apparatus 1, as in the present embodiment, the calculation unit 16 calculates the target range A0 so as to include the target position based on the thermal image. Such a configuration is preferable.

[0117] According to such a configuration, based on the thermal image, the target range A0 is calculated so as to include the target position. Thereby, the target range A0 can be set for the target thermal image.

[0118] Also, in the diagnostic apparatus 1, as in the present embodiment, the calculation unit 16 uses the target range setting thermal image P1a, which is the thermal image P1a with the target range A0 set, as teacher data, inputs the thermal image P2a, and outputs the target range calculation thermal image P3a, which is the thermal image P3a with the target range A0 calculated, to perform machine learning on the determination model. uses the determination model to output the target range calculation thermal image P3a from the specific thermal image P2a. Such a configuration is preferable.

[0119] According to such a configuration, the machine-learned determination model is a determination model that takes the thermal image P2a as input and outputs the target range calculation thermal image P3a. Then, using the determination model, the target range calculation thermal image P3a is output from the input thermal image P2a. Thereby, the target range A0 can be set for the target thermal image P2a.

[0120] Also, in the diagnostic apparatus 1, as in the present embodiment, the storage unit 15 stores a visible light image obtained by photographing the object with the visible light camera 6, the arithmetic unit 16 calculates the target range A0 so as to include the target position based on the thermal image and the visible light image, such a configuration is preferable.

[0121] According to such a configuration, based on the thermal image and the visible light image, the target range A0 is calculated so as to include the target position. Thereby, the target range A0 can be set for the target thermal image.

[0122] Also, in the diagnostic apparatus 1, as in the present embodiment, the arithmetic unit 16 divides the target range A0 into a first small range A1 including the target position and a second small range A2 determined to have an emissivity different from that of the first small range A1 based on the thermal image, calculates the emissivity of the first small range A1 based on the thermal image and the measured temperature, calculates the surface temperature of the first small range A1 based on the thermal image and the emissivity of the first small range A1, such a configuration is preferable.

[0123] According to such a configuration, based on the thermal image, the target range A0 is divided into a first small range A1 including the target position and a second small range A2 determined to have an emissivity different from that of the first small range A1. Then, based on the thermal image and the measured temperature, the emissivity of the first small range A1 is calculated, and based on the thermal image and the calculated emissivity, the surface temperature of the first small range A1 is calculated. Thereby, the surface temperature of the first small range A1 can be appropriately calculated.

[0124] Also, in the diagnostic apparatus 1, as in the present embodiment, the storage unit 15 stores a visible light image obtained by photographing the object with the visible light camera 6, the arithmetic unit 16 Based on the thermal image and the visible light image, the target range A0 is divided into a first small range A1 including the target position and a second small range A2 determined to have an emissivity different from that of the first small range A1. Based on the thermal image and the measured temperature, the emissivity of the first small range A1 is calculated. Based on the thermal image and the emissivity of the first small range A1, the surface temperature of the first small range A1 is calculated. Such a configuration is preferable.

[0125] According to such a configuration, based on the thermal image and the visible light image, the target range A0 is divided into a first small range A1 including the target position and a second small range A2 determined to have an emissivity different from that of the first small range A1. Then, based on the thermal image and the measured temperature, the emissivity of the first small range A1 is calculated, and based on the thermal image and the calculated emissivity, the surface temperature of the first small range A1 is calculated. Thereby, the surface temperature of the first small range A1 can be appropriately calculated.

[0126] Also, in the diagnostic apparatus 1, as in this embodiment, the calculation unit 16 calculates the surface temperature of the second small range A2 based on the surface temperature of the first small range A1. Such a configuration is preferable.

[0127] According to such a configuration, based on the calculated surface temperature of the first small range A1, the surface temperature of the second small range A2 is calculated. Thereby, the surface temperature of the second small range A2 can be appropriately calculated.

[0128] Also, the diagnostic method is, as in this embodiment, a diagnostic method for diagnosing an object using the above-described diagnostic apparatus 1, storing a thermal image of the object photographed by the thermal image camera 7, storing the emissivity of the target range A0 of the object, and calculating the surface temperature of the target range A0 based on the thermal image and the emissivity. The following method is preferred.

[0129] According to such a method, since the emissivity of the target range A0 of the object is taken into consideration, the surface temperature of the object can be appropriately calculated.

[0130] Also, the program, as in this embodiment, causes at least one processor to execute the above-described diagnostic method. The following configuration is preferred.

[0131] According to such a configuration, since the emissivity of the target range A0 of the object is taken into consideration, the surface temperature of the object can be appropriately calculated.

[0132] Note that the diagnostic apparatus 1, the diagnostic method, and the program are not limited to the configurations of the above-described embodiment, nor are they limited to the above-described effects. Also, the diagnostic apparatus 1, the diagnostic method, and the program can of course be variously modified without departing from the gist of the present invention. For example, it is of course possible to arbitrarily select one or more of the configurations and methods according to the following various modification examples and adopt them in the configurations and methods according to the above-described embodiment.

[0133] (A) In the diagnostic apparatus 1 according to the above-described embodiment, the calculation unit 16 calculates the emissivity of the target range A0 based on the thermal image and the measured temperature, and the storage unit 15 stores the calculated emissivity. However, the diagnostic apparatus 1 is not limited to such a configuration.

[0134] (A-1) For example, a configuration may be such that the emissivity of the target range A0 is input to the input unit 11, and the storage unit 15 stores the input emissivity. And, for example, the display unit 23a may display the surface temperature of the target position calculated from the thermal image, and the emissivity may be input to the input unit 11 so that the calculated surface temperature of the target position is the same as the measured temperature measured at the target position by the thermometer 3. Also, for example, the emissivity derived from the material or the like of the object may be input to the input unit 11.

[0135] Thus, the diagnostic apparatus 1 further includes an input unit 11 to which data is input, wherein the input unit 11 receives the emissivity of the target range A0, and the storage unit 15 stores the input emissivity. Such a configuration may be employed.

[0136] According to such a configuration, based on the thermal image of the object captured by the thermal imaging camera 7 and the emissivity of the target range A0 of the input object, the surface temperature of the target range A0 of the object is calculated. Thereby, since the emissivity of the target range A0 of the object is taken into account, the surface temperature of the object can be appropriately calculated.

[0137] (A-2) Further, for example, the material of the surface of the target range A0 (e.g., the material of the cover material 93) is input to the input unit 11, the storage unit 15 stores the relationship data between the material and the emissivity, and the calculation unit 16 calculates the emissivity of the target range A0 based on the input material and the stored relationship data, and the storage unit 15 stores the calculated emissivity. Such a configuration may be employed.

[0138] (B) Further, in the diagnostic apparatus 1 according to the above embodiment, the target position can be arbitrarily selected, and the target position (measurement position data) measured by the thermometer 3 is input to the input unit 21 (measurement position input unit 21e). However, the diagnostic apparatus 1 is not limited to such a configuration.

[0139] For example, the terminal device 2 may be configured to set the target position and display the set target position on the display unit 23a, that is, to measure the surface temperature of the target position set in the terminal device 2 with the thermometer 3. Although not particularly limited, the target position set by the terminal device 2 may be the position with the highest temperature among the thermal images captured by the thermal imaging camera 7. Thereby, for example, the temperature of the position where the heat insulation state of the pipe 9 is the worst can be accurately calculated.

[0140] (C) Further, in the diagnostic apparatus 1 according to the above embodiment, the storage unit 15 is configured to store the measured temperature obtained by measuring the surface of the target position within the target range A0 with the thermometer 3. However, the diagnostic apparatus 1 is not limited to such a configuration. For example, the diagnostic system 10 may be configured not to include the thermometer 3, and the storage unit 15 may not store the measured temperature.

[0141] (D) Further, in the diagnostic apparatus 1 according to the above embodiment, the calculation unit 16 is configured to calculate the target range A0 so as to include the target position based on the thermal image or based on the thermal image and the visible light image. However, the diagnostic apparatus 1 is not limited to such a configuration. For example, the position of the outer edge of the target range A0 may be continuously or intermittently input to the input unit 11, and the calculation unit 16 may calculate the target range A0 based on the position input to the input unit 11.

[0142] For example, the input unit 11 may be configured such that the position of the outer edge of the target range A0 is continuously or intermittently input with respect to the thermal image displayed on the display unit 13a. Then, the calculation unit 16 may be configured to calculate the target range A0 with respect to the thermal image based on the position input to the input unit 11.

[0143] Also, for example, the input unit 11 may be configured such that the position of the outer edge of the target range A0 is continuously or intermittently input with respect to the visible light image displayed on the display unit 13a. Then, the calculation unit 16 may be configured to calculate the position of the outer edge corresponding to the thermal image based on the position of the outer edge input to the input unit 11, and calculate the target range A0 with respect to the thermal image based on the calculated position corresponding to the thermal image.

[0144] (E) Further, in the diagnostic apparatus 1 according to the above embodiment, the calculation unit 16 is configured to divide the target range A0 into a first small range A1 and a second small range A2. However, the diagnostic apparatus 1 is not limited to such a configuration. For example, the calculation unit 16 may be configured not to divide the target range A0 into a plurality of small ranges A1, A2.

[0145] (F) Further, in the diagnostic apparatus 1 according to the above embodiment, the storage unit 15 is configured to store a visible light image obtained by photographing an object with the visible light camera 6. However, the diagnostic apparatus 1 is not limited to such a configuration. For example, the diagnostic system 10 may be configured not to include the visible light camera 6, and the storage unit 15 may be configured not to store a visible light image.

[0146] (G) Further, in the diagnostic apparatus 1 according to the above embodiment, the calculation unit 16 is configured to calculate the surface temperature of the second small range A2 based on the surface temperature of the first small range A1. However, the diagnostic apparatus 1 is not limited to such a configuration.

[0147] For example, the calculation unit 16 may be configured not to calculate the surface temperature of the second small range A2, that is, to exclude the second small range A2 from the target range A0. Further, for example, the emissivity of the second small range A2 may be input to the input unit 11, and the calculation unit 16 may be configured to calculate the surface temperature of the second small range A2 based on the thermal image and the input emissivity.

[0148] Further, for example, the material of the surface of the second small range A2 may be input to the input unit 11, the storage unit 15 may store relationship data between the material and the emissivity, and the calculation unit 16 may calculate the emissivity of the second small range A2 based on the input material and the stored relationship data, and further calculate the surface temperature of the second small range A2 based on the thermal image and the calculated emissivity.

[0149] (H) Further, in the diagnostic apparatus 1 according to the above embodiment, the visible light camera 6 and the thermal image camera 7 are configured to photograph an object at the same timing. However, the diagnostic apparatus 1 is not limited to such a configuration. For example, the visible light camera 6 and the thermal image camera 7 may be configured to photograph an object individually at different timings.

[0150] (I) Further, in the diagnostic apparatus 1 according to the above embodiment, after the target range calculation image P3 is calculated, the small range calculation image P6 is calculated. However, the diagnostic apparatus 1 is not limited to such a configuration.

[0151] For example, as shown in FIG. 12, the learning unit 16e uses the two-range setting image P7 in which the target range A0 and the small ranges A1 and A2 are set as teacher data, and uses the image P8 as an input to output the two-range calculation image P9 which is an image in which the target range A0 and the small ranges A1 and A2 are calculated. The two-range determination model may be machine-learned, and the target range calculation unit 16a may output the two-range calculation image P9 from a specific image P8 using the two-range determination model.

[0152] Although not particularly limited, for example, the target range calculation unit 16a may output the two-range calculation image P9 from a specific image P8 using a learned neural network. The model storage unit 15d may store, for example, the two-range determination model machine-learned by the learning unit 16e.

[0153] The combination of the two-range setting image P7, the image P8, and the two-range calculation image P9 is not particularly limited, and for example, the following first to third combinations and the like can be mentioned.

[0154] (I-1) As a first combination, for example, the two-range setting image P7 which is teacher data is a thermal image (two-range setting thermal image) P7a in which the target range A0 and the small ranges A1 and A2 are set, the input image P8 is a thermal image P8a, and the output two-range calculation image P9 is a thermal image (two-range calculation thermal image) P9a in which the target range A0 and the small ranges A1 and A2 are calculated. Such a configuration may be used.

[0155] In such a configuration, the target range calculation unit 16a outputs the two-range calculation thermal image P9a from the thermal image P8a using the two-range determination model. Thereby, the target range calculation unit 16a calculates the target range A0 and the small ranges A1 and A2 based on the thermal image P8a.

[0156] (I-2) Further, as a second combination, for example, the two-range setting image P7 which is teacher data is a visible light image (two-range setting visible light image) P7b in which the target range A0 and the small ranges A1, A2 are set, the input image P8 is a visible light image P8b, and the output target range calculation image P9 may be a visible light image (two-range calculation visible light image) P9b in which the target range A0 and the small ranges A1, A2 are calculated.

[0157] In such a configuration, the target range calculation unit 16a outputs a two-range calculation visible light image P9b from the visible light image P8b using a two-range determination model, and then the target range calculation unit 16a calculates a two-range calculation thermal image P9a based on the thermal image 8a (capturing the pipe 9 in the same area) associated with the visible light image P8b and the two-range calculation visible light image P9b. Therefore, the target range calculation unit 16a calculates the target range A0 and the small ranges A1, A2 based on the associated thermal image P8a and visible light image P8b.

[0158] (I-3) Further, as a third combination, for example, the two-range setting image P7 which is teacher data is a two-range setting thermal image P7a and a two-range setting visible light image P7b associated with each other (capturing the pipe 9 in the same area), the input image P8 is a thermal image P8a and a visible light image P8b associated with each other (capturing the pipe 9 in the same area), and the output two-range calculation image P9 may be a two-range calculation thermal image P9a and a two-range calculation visible light image P9b.

[0159] In such a configuration, the target range calculation unit 16a outputs a two-range calculation thermal image P9a and a two-range calculation visible light image P9b from the associated thermal image P8a and visible light image P8b, for example, using a two-range determination model. Thereby, the target range calculation unit 16a calculates the target range A0 and the small ranges A1, A2 based on the associated thermal image P8a and visible light image P8b.

[0160] (J) Further, for example, at least a part of the components of each part 21, 22, 23 of the terminal device 2 according to the above embodiment may be provided in each part 11, 12, 13 of the diagnostic device 1. Also, for example, at least a part of the components of each part 11, 12, 13 of the diagnostic device 1 according to the above embodiment may be provided in each part 21, 22, 23 of the terminal device 2.

[0161] Although not particularly limited, for example, the measured temperature input unit 21d may be provided in the input unit 11 of the diagnostic device 1, and also, for example, the measured position input unit 21e may be provided in the input unit 11 of the diagnostic device 1. Also, for example, the input unit 11 of the diagnostic device 1 may include an image input unit to which the captured images (visible light image, thermal image) are input.

[0162] As described above, the input unit of the diagnostic system 10 is composed of the input unit 11 of the diagnostic device 1 and the input unit 21 of the terminal device 2, the processing unit of the diagnostic system 10 is composed of the processing unit 12 of the diagnostic device 1 and the processing unit 22 of the terminal device 2, and the output unit of the diagnostic system 10 is composed of the output unit 13 of the diagnostic device 1 and the output unit 23 of the terminal device 2. Note that, for example, the diagnostic system 10 may be configured to include only the diagnostic device 1 without including the terminal device 2.

[0163] Also, for example, the diagnostic device 1 according to the above embodiment is configured to execute the target range calculation step S3 to the heat retention state diagnosis step S7 among the steps S1 to S7 of the diagnostic method. However, the diagnostic device 1 is not limited to such a configuration. For example, the diagnostic device 1 may be configured to execute the imaging step S1 to the surface temperature calculation step S6 among the steps S1 to S7 of the diagnostic method, or may be configured to execute all the steps S1 to S7 of the diagnostic method.

[0164] (J-1) Here, the diagnostic device 1 that executes the photographing process S1 to the surface temperature calculation process S6 will be described with reference to FIGS. 13 to 15. First, the configuration of the diagnostic device 1 will be described. The following configuration is an example for assisting in understanding the configuration of the diagnostic device 1 and does not limit the configuration of the diagnostic device 1.

[0165] As shown in FIG. 13, the diagnostic device 1 may be, for example, a tablet computer. Then, as shown in FIGS. 13 and 14, the diagnostic device 1 may include, for example, a visible light camera 6 and a thermal image camera 7, an input unit (device input unit) 11, a processing unit (device processing unit) 12, and an output unit (device output unit) 13.

[0166] The device input unit 11 includes, for example, a touch panel 11a, buttons 11b, a photographing instruction input unit 11c, a measured temperature input unit 11d, and a measurement position input unit 11e. The device output unit 13 includes, for example, a display unit (display) 13a, a sound output unit 13b, and a transmission unit (not shown).

[0167] As shown in FIG. 14, the device storage unit 15 includes, for example, an image storage unit 15a, a temperature storage unit 15b, an emissivity storage unit 15c, and a model storage unit 15d. The device calculation unit 16 includes, for example, a target range calculation unit 16a, an emissivity calculation unit 16b, a surface temperature calculation unit 16c, and a learning unit 16e.

[0168] Next, the diagnostic method (photographing process S1 to surface temperature calculation process S6) executed by the diagnostic device 1 will be described. The following diagnostic method is an example for assisting in understanding the diagnostic method of the diagnostic device 1 and does not limit the diagnostic method of the diagnostic device 1.

[0169] First, the diagnostic device 1 executes a photographing process S1. In the photographing process S1, for example, by inputting once of photographing instruction data to the photographing instruction input unit 11c, each of the visible light camera 6 and the thermal image camera 7 photographs the pipe 9 simultaneously or continuously. Thereby, the visible light image photographed by the visible light camera 6 and the thermal image photographed by the thermal image camera 7 can be imaged without a time difference so that they can be easily corresponded. Then, for example, as shown in FIG. 15, the display unit 13a of the diagnostic device 1 displays the visible light image (the left image in FIG. 15) of the pipe 9 photographed by the visible light camera 6 and the thermal image (the right image in FIG. 15) of the pipe 9 photographed by the thermal image camera 7.

[0170] Note that the device arithmetic unit 16 may be configured to process the visible light image and the thermal image so that, for example, the sizes of the objects in both images are the same (for example, trimming, enlarging, or reducing). In such a configuration, for example, the device arithmetic unit 16 may be configured to extract the feature points of both images and compare or match them so as to process the visible light image and the thermal image so that the sizes of the objects in both images are the same. Also, for example, the device arithmetic unit 16 may be configured to process the visible light image and the thermal image so as to match the positions and sizes of the feature points of the visible light image and the thermal image.

[0171] Next, the diagnostic device 1 executes a temperature measurement process S2. In the temperature measurement process S2, for example, the surface temperature of the target position of the pipe 9 is measured by the thermometer 3. Then, for example, the measured temperature data is input to the measured temperature input unit 11d, and the measured position data is input to the measured position input unit 11e. Thereby, for example, as shown in FIG. 15, the display unit 13a of the terminal device 2 displays the surface temperature of the target position of the pipe 9 (the temperature displayed as "measured temperature" in FIG. 15) and the position of the target position (the position of the "+" in the left thermal image in FIG. 15).

[0172] For example, as shown in FIG. 15, when the position of the target position (in FIG. 15, the position of "×" in the visible light image on the left side) is input to the measurement position input unit 11e with respect to the visible light image displayed on the display unit 13a, the device arithmetic unit 16 may be configured to calculate the position of the thermal image corresponding to the position of the input target position (in FIG. 15, the position of "+" in the thermal image on the right side).

[0173] Thereby, the target position measured by the thermometer 3 can be input with respect to the visible light image. Therefore, for example, the target position measured by the thermometer 3 can be easily input. Note that, for example, when correction instruction data (for example, the position with respect to the thermal image) is input to the measurement position input unit 11e, the device arithmetic unit 16 may correct the calculated position of the thermal image. Then, the device storage unit 15 stores the visible light image and the thermal image captured in the imaging step S1 in association with the measured temperature of the target position and the position of the target position input in the temperature measurement step S2.

[0174] Next, the diagnostic device 1 executes a target range calculation step S3. Although not particularly limited, in the target range calculation step S3, for example, the device input unit 11 continuously or intermittently inputs the position of the outer edge of the target range A0 with respect to the thermal image or the visible light image displayed on the display unit 13a, and the device arithmetic unit 16 calculates the target range A0 with respect to the thermal image based on the position input to the input unit 11, and the display unit 13a may be configured to display the target range A0 on the thermal image.

[0175] Next, the diagnostic device 1 executes a small range calculation step S4. Note that the diagnostic device 1 does not necessarily have to execute the small range calculation step S4.

[0176] Next, the diagnostic device 1 executes an emissivity calculation step S5. Although not particularly limited, in the emissivity calculation step S5, for example, the emissivity of the target position is input to the device input unit 11, the device storage unit 15 stores the input emissivity, and the device calculation unit 16 calculates the temperature of the target position based on the thermal image (specifically, the thermal data of the target position) and the emissivity input to the input unit 11, and the display unit 13a may display the calculated temperature.

[0177] Thereby, for example, when a specific emissivity is input to the device input unit 11 and the temperature of the target position calculated in the emissivity calculation step S5 is the same as the temperature of the target position measured in the temperature measurement step S2, the specific emissivity can be stored in the device storage unit 15 as the emissivity of the target position (specifically, the emissivity of the target range A0 including the target position).

[0178] Next, the diagnostic device 1 executes a surface temperature calculation step S6. Although not particularly limited, in the surface temperature calculation step S6, for example, the device calculation unit 16 calculates the temperature of the target range A0 based on the thermal image (specifically, the thermal data of the target range A0) and the specific emissivity stored in the measurement storage unit 15 in the emissivity calculation step S5, and the display unit 13a may display the calculated temperature (for example, the average temperature of the target range A0).

[0179] Thereafter, the diagnostic device 1 outputs, for example, the data stored and calculated in each of the steps S1 to S6 to a downstream device (for example, a desktop personal computer, a notebook personal computer, etc., not shown), and the downstream device stores the data output from the diagnostic device 1. Then, the downstream device executes, for example, a heat retention state diagnosis step S7.

[0180] Thus, in the diagnostic device 1, with such a configuration, the storage unit 15 stores a visible light image obtained by photographing the object with the visible light camera 6 and a measured temperature obtained by measuring the surface of the target position within the target range A0 with the thermometer 3. the diagnostic device 1 A display unit 13a that displays the thermal image and the visible light image, an input unit 11 for inputting the position of the target position with respect to the visible light image displayed on the display unit 13a, The arithmetic unit 16 calculates the position of the thermal image corresponding to the position of the target position input to the input unit 11 based on the thermal image and the visible light image. Such a configuration is preferable.

[0181] According to such a configuration, when the position of the target position is input with respect to the visible light image displayed on the display unit 13a, the position of the thermal image corresponding to the position of the target position input to the input unit 11 is calculated based on the thermal image and the visible light image. Thereby, the target position measured by the thermometer 3 can be specified by the visible light image.

[0182] Further, in the diagnostic apparatus 1, as in this configuration, the display unit 13a displays the position calculated by the arithmetic unit 16 on the thermal image. Such a configuration is preferable.

[0183] According to such a configuration, since the calculated position is displayed on the thermal image by the display unit 13a, the calculated position can be easily confirmed.

[0184] (K) Note that, for example, in the claims, the specification, and the drawings, the execution order of each process such as operations, procedures, steps, and stages in the system, method, program, and apparatus shown can be realized in any order as long as the output of the previous process is not used in the subsequent process. For example, even if described using "first", "next", etc. for convenience, it does not mean that it is essential to execute in this order.

Explanation of Reference Numerals

[0185] 1... Diagnostic device, 2... Terminal device, 2a... Main body, 2b... Cable, 3... Thermometer, 4... Memory device, 5... Communication means, 6... Visible light camera, 7... Thermal imaging camera, 9... Pipe, 10... Diagnostic system, 11... Device input section, 11a... Touch panel, 11b... Button, 11c... Shooting instruction input section, 11d... Measured temperature input section, 11e... Measurement position input section, 12... Device processing section, 13... Device output section, 13a... Display section, 13b... Sound output section, 14... Device acquisition section, 15... Device memory section, 15a... Image memory section, 15b... Temperature memory section, 15c... Emissivity memory section, 15d... Model memory section, 16... Device calculation section, 16a... Target range calculation section, 16b... Emissivity calculation section, 16c... Surface temperature calculation section, 16d... Heat insulation state diagnosis section, 16e... Learning section, 17... Device control section, 21... Terminal input section, 21a... Touch panel, 21b... Button, 21c... Shooting instruction input section, 21d... Measured temperature input section, 21e... Measurement position input section, 22... Terminal processing section, 23... Terminal output section, 23a... Display section, 23b... Sound output section, 24... Terminal acquisition section, 25... Terminal memory section, 26... Terminal calculation section, 27... Terminal control section, 91... Pipe material, 92... Heat insulation material, 93... Cover material, A0... Target range, A1... First small range, A2... Second small range, P1... Target range setting image, P1a... Target range setting thermal image, P1b... Target range setting visible light image, P2... Image, P2a... Thermal image, P2b... Visible light image, P3... Target range calculation image, P3a... Target range calculation thermal image, P3b... Target range calculation visible light image, P4... Small range setting image, P4a... Small range setting thermal image, P4b... Small range setting visible light image, P5... Image, P5a... Thermal image, P5b... Visible light image, P6... Small range calculation image, P6a... Small range calculation thermal image, P6b... Small range calculation visible light image, P7... Both range setting image, P7a... Both range setting thermal image, P7b... Both range setting visible light image, P8... Image, P8a... Thermal image, P8b... Visible light image, P9... Both range calculation image, P9a... Both range calculation thermal image, P9b... Both range calculation visible light image

Claims

1. A diagnostic device for diagnosing an object, comprising: A storage unit that stores a thermal image of the object captured by a thermal imaging camera and the emissivity of a target area of ​​the object; a calculation unit that calculates a surface temperature of the target area based on the thermal image and the emissivity.

2. The memory unit stores a measured temperature measured by a temperature measuring device on a surface at a target position within the target range, The calculation unit calculates an emissivity of the target area based on the thermal image and the measured temperature, The diagnostic device according to claim 1 , wherein the storage unit stores the calculated emissivity.

3. The diagnostic device according to claim 2 , wherein the calculation unit calculates the target range based on the thermal image so as to include the target position.

4. The calculation unit is Using a target range set thermal image, which is a thermal image in which the target range is set, as teacher data, a judgment model is machine-learned to input the thermal image and output a target range calculated thermal image, which is a thermal image in which the target range is calculated; The diagnostic device according to claim 3 , further comprising: a determining unit configured to determine a range of thermal images from the specific thermal images using the determining model;

5. The storage unit stores a visible light image of the object captured by a visible light camera, The diagnostic device according to claim 2 , wherein the calculation unit calculates the target range so as to include the target position based on the thermal image and the visible light image.

6. The calculation unit is Based on the thermal image, the target range is divided into a first small range including the target position and a second small range determined to have an emissivity different from the emissivity of the first small range; Calculating the emissivity of the first small range based on the thermal image and the measured temperature; The diagnostic device according to claim 3 , further comprising a step of calculating a surface temperature of the first small area based on the thermal image and the emissivity of the first small area.

7. The storage unit stores a visible light image of the object captured by a visible light camera, The calculation unit is Based on the thermal image and the visible light image, the target range is divided into a first small range including the target position and a second small range determined to have an emissivity different from the emissivity of the first small range; Calculating the emissivity of the first small range based on the thermal image and the measured temperature; The diagnostic device according to claim 3 , further comprising: a surface temperature of the first small area calculated based on the thermal image and the emissivity of the first small area.

8. The calculation unit is Based on the thermal image and the visible light image, the target range is divided into a first small range including the target position and a second small range determined to have an emissivity different from the emissivity of the first small range; Calculating the emissivity of the first small range based on the thermal image and the measured temperature; The diagnostic device according to claim 5, further comprising a step of calculating a surface temperature of the first small area based on the thermal image and the emissivity of the first small area.

9. 9. The diagnostic device according to claim 6, wherein the calculation unit calculates the surface temperature of the second small area based on the surface temperature of the first small area.

10. Further comprising an input unit for inputting data, The input unit receives an input of the emissivity of the target range, 9. The diagnostic device according to claim 1, wherein the storage unit stores the input emissivity.

11. The storage unit stores a visible light image of the object captured by a visible light camera and a measured temperature of a surface of the object position within the object range measured by a temperature measuring instrument; The diagnostic device comprises: a display unit for displaying the thermal image and the visible light image; an input unit to which a position of the target position is input with respect to the visible light image displayed on the display unit, The diagnostic device according to claim 1 , wherein the calculation unit calculates a position of the thermal image corresponding to a position of the target position input to the input unit, based on the thermal image and the visible light image.

12. The diagnostic device according to claim 11 , wherein the display unit displays the position calculated by the calculation unit on the thermal image.

13. A diagnostic method for diagnosing an object using the diagnostic device according to any one of claims 1 to 8, comprising: storing a thermal image of the object captured by a thermal imaging camera; storing the emissivity of a target area of ​​the target; and calculating a surface temperature of the target area based on the thermal image and the emissivity.

14. A program causing at least one processor to execute the diagnostic method according to claim 13.

Images