JP2018141778A - Method of deriving apparent emissivity, temperature measurement method, method of manufacturing pipe materials, and temperature measurement device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method of deriving apparent emissivity, a method of measuring temperature, a method of manufacturing pipe materials, and a temperature measurement device, which allow for accurately measuring temperature of a pipe-end inner surface in consideration of the influence of multiple reflections.SOLUTION: A method of deriving apparent emissivity assumes light volume obtained at a pipe-end inner surface, or an inner surface of a tubular object at an end thereof, to be a sum of a self-emitted light component directly emitted from the pipe-end inner surface and a multiple reflection component which is emission light that has gone through multiple reflections on the pipe inner surface along an axial direction of the tubular object before being emitted from the pipe-end inner surface in order to calculate apparent emissivity of the pipe-end inner surface.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an apparent emissivity calculation method, a temperature measurement method, a pipe material manufacturing method, and a temperature measurement device.

  In the manufacturing process of steel pipe, which is a pipe material, the temperature information of the steel pipe is a very important factor in operation. Even if the steel pipe is manufactured with the same composition and rolling load, the temperature conditions are different, so Large differences in material may occur. The temperature condition described here includes not only the temperature of the outer surface of the steel pipe but also the temperature of the inner surface, and the temperature condition of the inner surface heated by the processing heat generation is particularly important. Therefore, conventionally, there has been a temperature measuring method for measuring the temperature of the inner surface of the tube by assuming the emissivity of the inner surface of the steel tube in advance and converting the luminance (light quantity) of the inner surface of the tube obtained by a camera as an imaging means into temperature. Are known.

  Patent Document 1 discloses a camera for capturing an image representing thermal radiance at a tube end including a tube inner surface of a steel tube, an image processing device for processing the image and recognizing the tube inner surface, and image processing. There is disclosed a temperature measuring device including an arithmetic unit for calculating the temperature of a tube inner surface using the thermal radiance signal of the tube inner surface sent from the device and the recognition information of the tube inner surface. In this temperature measurement device, the thermal radiance signal of the inner surface portion of the tube is regarded as having an effective emissivity of 1, and the temperature of the inner surface portion of the tube is calculated by the arithmetic unit.

Japanese Patent Laid-Open No. 5-231949

Iuchi et al., “Non-contact temperature measurement of metals during cold and hot rolling”, Amada Metalworking Machine Technology Promotion Foundation Research Summary Report, International Exchange Report 12th, 2000, p.123-129 Ito et al., “Estimation of Light Source Environment Based on Object Shading”, IPSJ Transactions on Computer Vision and Image Media, vol.41 No.SIG 10, Dec. 2000 JIS C 1612: 2000 “General Rules for Performance Tests of Radiation Thermometers”

  However, in practice, the emissivity of the inner surface of the tube is not simply 1, and a complex multiple reflection phenomenon occurs due to the inner surface structure of the steel tube, so that it is difficult to calculate the apparent emissivity. The disclosed temperature measuring device has a problem that the temperature of the inner surface of the tube cannot be measured with high accuracy.

  The present invention has been made in view of the above problems, and its purpose is to provide an apparent radiation that enables accurate measurement of the temperature of the inner surface of the tube at the tube end in consideration of the influence of multiple reflection. A rate calculation method, a temperature measurement method, a pipe material manufacturing method, and a temperature measurement device are provided.

  In order to solve the above-described problems and achieve the above object, an apparent emissivity calculation method according to the present invention is configured to obtain a light amount obtained from a tube end inner surface which is a tube inner surface portion of a tube end portion of a tube-shaped object. The tube as a sum of a self-luminous component directly emitted from the inner surface of the end and a multiple reflection component of the reflected light emitted from the inner surface of the tube by multiple reflection over the axial direction of the tube-shaped object. The apparent emissivity of the end inner surface is calculated.

  Further, the temperature measurement method according to the present invention is a temperature measurement method for measuring the temperature of the tube end inner surface which is the tube inner surface portion at the tube end portion of the tube-shaped object, and the image of the tube end portion including the tube end inner surface. From the image processing step for recognizing the tube end inner surface, the luminance value of the tube end inner surface recognized in the image processing step, and the tube end calculated by the apparent emissivity calculation method of the invention And a temperature calculating step for calculating the temperature of the inner surface of the tube end using the apparent emissivity of the inner surface.

  The temperature measurement method according to the present invention is a temperature measurement method for measuring the temperature of the inner surface of the tube end, which is the inner surface of the tube at the tube end of the tube-shaped object, and the light receiving angle of the imaging means for imaging the tube end. θ is 60 ° or less, the inner diameter of the tube-shaped object is d, the length of the tube-shaped object in the axial direction is L, and one of the parameters representing the surface properties of the inner surface of the tube is σ. In this case, the following relationship (1) is satisfied. The unit of σ described here is [°].

  In the temperature measurement method according to the present invention as set forth in the invention described above, the light receiving angle θ recognizes the tube end surface or the tube end inner surface at the tube end from the image of the tube end captured by the imaging means. And it calculates using the cross-sectional shape in the direction orthogonal to the said axial direction of the said tube end surface or the said tube end inner surface.

  Moreover, the manufacturing method of the pipe material which concerns on this invention manufactures the pipe material which is the said pipe-shaped object using the information of the temperature of the said pipe end inner surface measured using the temperature measuring method of said invention, It is characterized by the above-mentioned. To do.

  The temperature measuring device according to the present invention is a temperature measuring device that measures the temperature of the inner surface of the tube end that is the inner surface of the tube at the tube end of the tube-shaped object, and the tube end portion includes the inner surface of the tube end. An image processing unit that performs image processing for recognizing the inner surface of the tube end from an image captured by the imaging unit, and a luminance value of the inner surface of the tube end recognized by the image processing And an apparent emissivity of the inner surface of the tube end calculated by the apparent emissivity calculation method of the invention described above, and a temperature calculation means for calculating the temperature of the inner surface of the tube end. It is what.

  The apparent emissivity calculation method, temperature measurement method, tube material manufacturing method, and temperature measurement apparatus according to the present invention can accurately measure the temperature of the inner surface of the tube end in consideration of the influence of multiple reflection. There is an effect.

Drawing 1 is a mimetic diagram of a temperature measuring device for measuring temperature of a pipe inner surface part (tube end inner surface) in a pipe end part of a steel pipe concerning an embodiment. FIG. 2 is a schematic diagram of an experimental apparatus used for calculating the surface emissivity of a steel material sample. FIG. 3 is a graph showing the radiation angle characteristics of the steel material. FIG. 4 is an explanatory diagram of the light receiving angle θ of the area sensor. Fig.5 (a) is a figure which shows distribution of the surface emissivity of a steel pipe. FIG.5 (b) is a figure which shows the image which imaged the red hot steel pipe from the side with the area sensor. FIG. 6 is a graph showing the radiation angle dependency. FIG. 7 is a diagram showing an outline of a simulation of multiple reflection that occurs at the inner surface of a steel pipe. FIG. 8 is a graph showing the acceptance angle characteristic of the apparent emissivity when the reflectance is changed with the ratio L / d = 10 of the tube length L and the tube diameter d. FIG. 9 is a graph showing the results of evaluating the acceptance angle characteristics of the apparent emissivity when the emissivity is 0.9, separately for radiation and multiple reflection effects. FIG. 10 is a graph when the ratio L / d is changed for a highly specular surface. FIG. 11 is a graph when the ratio L / d is changed for a surface with low specularity. FIG. 12A is an explanatory diagram of the apparent emissivity when the tube inner surface portion exists in the regular reflection direction. FIG. 12B is an explanatory diagram of the apparent emissivity when the tube inner surface portion does not exist in the regular reflection direction. FIG. 13 is an explanatory diagram of the temperature distribution in the steel pipe axial direction when the temperature is varied at both ends in the steel pipe axial direction. FIG. 14 is a graph showing the results when the light receiving angle θ is changed when the temperature is varied at both ends of the steel pipe axis direction. Fig.15 (a) is a figure which shows the image of the pipe end part containing the pipe | tube inner surface part of a steel pipe. FIG. 15B is a diagram showing a histogram generated from the image of FIG. FIG. 16 shows a tube end image, a brightness profile in the longitudinal direction of the tube diameter center portion generated from the image, and an emissivity calculated by simulation for light reception angles θ of 45 [°] and 60 [°]. It is a figure. FIG. 17 shows a tube end image, a luminance profile in the longitudinal direction of the tube diameter center generated from the image, and the presence or absence of a flat portion having a substantially constant luminance and a flatness in the luminance profile. It is the figure shown about (degree), 50 [degree], 55 [degree], and 60 [degree]. FIG. 18 is a brightness profile in the longitudinal direction of the center of the tube diameter, which is generated from a tube end image, taken by changing the thickness t and the light receiving angle θ of the steel tube on an actual rolling line. FIG. 19 is a diagram illustrating a tube end image and a tube inner surface portion, a flat portion, and a tube outer surface portion in a luminance profile in the longitudinal direction of the tube diameter center portion generated from the image. FIG. 20 is an explanatory diagram of a method of calculating the length of the tube inner surface portion and the length of the flat portion in the luminance profile. FIG. 21 is a histogram of indices corresponding to the luminance profiles in FIG. FIG. 22 is a graph showing the average and standard deviation of the index at each light receiving angle θ and each thickness t corresponding to each luminance profile in FIG. FIG. 23 is a diagram illustrating a simulation result for each light receiving angle θ when the emissivity is 0.8. FIG. 24 is a diagram showing an arrangement position of the area sensor with respect to the steel pipe. FIG. 25 is a diagram illustrating an image of a pipe end portion including a pipe inner surface portion of a steel pipe.

  Hereinafter, an embodiment of an apparent emissivity calculation method, a temperature measurement method, a pipe material manufacturing method, and a temperature measurement device according to the present invention will be described. In addition, this invention is not limited by this embodiment.

  Drawing 1 is a mimetic diagram of temperature measuring device 1 for measuring the temperature of pipe inner surface part 12 (tube end inner face) in pipe end part 11 of steel pipe 10 concerning an embodiment. The temperature measuring device 1 is composed of an area sensor 2, an image processing device 3, an arithmetic device 4, and the like that are imaging means. The area sensor 2 may be provided with a wavelength selection filter such as an interference filter in order to limit the wavelength. The visual field of the area sensor 2 is set so that the tube inner surface 12 at the tube end 11 enters the screen, and not only the self-luminous component directly emitted from the tube inner surface 12 at the tube end 11 but also the steel tube 10. The radiance is obtained by adding the multiple reflection components obtained by multiple reflections of the radiation from other portions of the tube inner surface portion 12 along the axial direction. In the present embodiment, as shown in FIG. 1, the area sensor 2 receives an angle formed by the optical axis X of the area sensor 2 and a virtual straight line Y orthogonal to the steel pipe axis direction from the pipe end portion 11. In this case, the light receiving angle θ is set. In addition, the pipe end part 11 described here refers to the inner surface of the pipe end part 11 by the mesh width of the simulation described later. The obtained radiance is used for evaluating the influence of multiple reflections using the emissivity measured under the condition that the surface property of the tube inner surface portion 12 is not subjected to multiple reflection (hereinafter referred to as surface emissivity). Using the emissivity calculated by simulation (hereinafter, defined as apparent emissivity), the calculation device 4 converts the emissivity into temperature.

  The temperature measuring method used in the temperature measuring apparatus 1 according to the present embodiment is configured by the following three processing steps, the first processing step to the third processing step. In the first treatment step, the surface emissivity and the reflectance are measured for the surface properties of the tube inner surface portion 12. In the second processing step, the radiance of the tube inner surface portion 12 at the tube end portion 11 of the steel tube 10 is estimated by simulation, and a condition capable of stably measuring the temperature is derived. In the third processing step, using the conditions derived in the second processing step, a thermometer is actually installed and temperature measurement is performed.

  First, the first treatment process in the temperature measurement method according to the present embodiment will be described. In order to simulate the reflection of the inner surface and evaluate the influence of multiple reflections, the physical characteristics of radiation and reflection on the steel surface are required in advance. First, the surface emissivity is calculated.

  FIG. 2 is a schematic diagram of the experimental apparatus 20 used for calculating the surface emissivity of the steel material sample 22. The experimental apparatus 20 includes an area sensor 2, a heating furnace 21, a steel material sample 22, a brick 23, a thermocouple 24, a logger 25, and the like. An opening 21a such as a vent is opened on the upper surface of the heating furnace 21, and the area sensor 2 is arranged so as to look into the heating furnace 21 from the opening 21a. In the heating furnace 21, the steel material sample 22 is installed on the brick 23 so as to be heated by radiant heat from a heater (not shown) provided in the vicinity of the inner surface of the heating furnace 21. The steel material sample 22 is located in the upper part in the heating furnace 21 as much as possible so that it may not receive. In the steel sample 22, thermocouples 24 are embedded at a plurality of depth positions near the top surface of the steel sample (the surface viewed by the area sensor 2), and the output value from the thermocouple 24 is recorded by the logger 25. The surface temperature of the steel material sample 22 is calculated by extrapolating using a two-dimensional heat transfer model (linear model). The area sensor 2 limits the wavelength according to the temperature range, calibrates beforehand with a black body furnace, and measures the surface emissivity based on the surface temperature measured and calculated using the thermocouple 24 and the radiance at that time. calculate. In addition, by applying a black body spray that can artificially make the surface close to a black body on a part of the upper surface of the steel sample 22, the luminance is compared between the part where the black body spray is applied and the part where the black body spray is not applied. The surface emissivity may be calculated.

  Next, the angle characteristic of the surface emissivity of the steel pipe 10 is derived. Conventionally, it is known that there is an angle dependency as shown in FIG. 3 with respect to a radiation angle characteristic of a steel material used for manufacturing the steel pipe 10. That is, as shown in FIG. 4, when the light receiving angle θ of the area sensor 2 is increased with respect to the vertical direction orthogonal to the axial direction of the steel pipe 10, the emissivity changes. Further, the surface property in a uniform and thermal equilibrium state is measured by experiment by changing the light receiving angle θ of the area sensor 2 and obtaining the luminance with respect to the sample radiation surface equivalent to the tube inner surface portion 12 of the steel tube 10. Also good.

  In addition, the emissivity needs to make the emissivity at the time of the perpendicular | vertical radiation which is a radiation | emission to a perpendicular direction with respect to the steel material surface correspond with the above-mentioned surface emissivity. In this embodiment, the distribution (histogram) of the surface emissivity of the steel pipe 10 as shown in FIG. 5 (a), and the red hot steel pipe 10 is horizontally (steel pipe) by the area sensor 2 as shown in FIG. 5 (b). The angle characteristic of the surface emissivity was derived from an image captured from a direction orthogonal to the ten axial directions and used in a simulation described later. FIG. 6 is a graph showing the radiation angle dependence, where the vertical axis shows the relative emissivity when the vertical radiation is 1, and the horizontal axis is the angle from the vertical direction. A certain light receiving angle θ is shown. This is qualitatively consistent with FIG. 3 described in Non-Patent Document 1.

Next, the angle characteristic of reflectance is derived. In order to simulate the multiple reflection occurring at the pipe inner surface portion 12 of the steel pipe 10, it is necessary to physically evaluate the reflection characteristics with respect to the incident angle and the reflection angle. Therefore, evaluation is performed using a simplified model (see Non-Patent Document 2) of the simplified Torrance-Sparrow model shown in the following equation (2), where the incident angle is θ _i and the reflection angle is θ _r . The Torrance-Sparrow model is a model that expresses reflection as the sum of specular reflection and diffuse reflection assuming a small surface element. K in the following equation (2) is a reflection parameter representing the contribution of each reflection and the spread of specular reflection (representing the surface properties), and the parameters K ₁ , K ₂ and σ in the following equation (2) Is a three-dimensional vector. Therefore, in this specification, k = (K ₁ , K ₂ , σ) is also expressed. This k is obtained by measuring the reflection distribution with respect to the incident angle by experiment and fitting the following equation (2). In this embodiment, k = (K ₁ , K ₂ , σ) = (800, 200, 13.2). The reflectance described here is normalized by the sum of the reflectances calculated from the angular characteristics of surface radiation and Kirchhoff's law.

Next, the second processing step in the temperature measurement method according to the present embodiment will be described. FIG. 7 is a diagram showing an outline of a simulation of multiple reflection occurring at the pipe inner surface portion 12 of the steel pipe 10. In FIG. 7, L is the pipe length (length in the axial direction of the steel pipe 10), and d is the pipe diameter (inner diameter of the steel pipe 10). First, the steel pipe 10 is divided into M regions in the axial direction. Next, the amount of light beam L _{p + 1} (m _s , _me ) from the m _s th region to the _me th region at the time of p + 1 reflection (p = 0, 1, 2, 3,...) Is p times m th in reflection (m = 1 to m) light incident on the area of the _{m s} th _(m s = 1 to m) from each region of the area of each _{m e} th _(m e = 1 to m) Using the reflectance α (m, m _s , m _e ) at the time of reflection toward the surface, it can be expressed by the following equation (3).

For example, when the steel pipe 10 is divided into six (M = 6) regions in the axial direction, the fifth from the second (m _s = 2) region at the time of reflection 5 times (p + 1 = 5 on the left side) The amount of light beam L ₅ (2, 5) going to the area of (m _e = 5) is the light incident on the second area from each of the first to sixth areas when reflected four times (p = 4 on the right side) Using the reflectance α (m, 2, 5) when reflecting toward the fifth region, it can be expressed by the following equation (4).

In the above equation (4), L ₄ (1,2) represents the amount of light incident on the second region from the first region during four reflections. L ₄ (2, 2) indicates the amount of light incident on the second region from the second region upon four reflections. L ₄ (3, 2) indicates the amount of light incident on the second region from the third region upon four reflections. L ₄ (4, 2) indicates the amount of light incident on the second region from the fourth region upon four reflections. L ₄ (5, 2) indicates the amount of light incident on the second region from the fifth region upon four reflections. L ₄ (6, 2) indicates the amount of light incident on the second region from the sixth region upon four reflections.

  In the above equation (4), α (1, 2, 5) is a reflection when light incident on the second region from the first region is reflected toward the fifth region at the next reflection. Rate. Α (2, 2, 5) is a reflectance when light incident on the second region from the second region is reflected toward the fifth region at the next reflection. Α (3, 2, 5) is a reflectance when light incident on the second region from the third region is reflected toward the fifth region at the next reflection. Further, α (4, 2, 5) is a reflectance when light incident on the second region from the fourth region is reflected toward the fifth region at the time of the next reflection. Α (5, 2, 5) is a reflectance when light incident on the second region from the fifth region is reflected toward the fifth region at the time of the next reflection. Α (6, 2, 5) is a reflectance when light incident on the second region from the sixth region is reflected toward the fifth region at the next reflection.

Next, the incident angle θ _i can be expressed by the following equation (5) using the tube length L, the tube diameter d, and the mesh M. Further, the reflection angle θ _r can be expressed by the following equation (6) using the tube length L, the tube diameter d, and the mesh M. Therefore, the reflectance α (m, m _s , m _e ) is calculated by adding the incidence angle θ _i and the reflection angle θ _{r of the following} formula (5) and the following formula (6) to the Torrance-Sparrow model shown in the above formula (2). Can be obtained by substituting. The reflection parameter k = (K ₁ , K ₂ , σ) was derived by experiment. The reflectance _{α (m, m s, m} e) , based on the Kirchhoff's law, were normalized to a 1 by adding the sum of reflectance and emissivity using the following equation (7). Further, the emissivity used for this normalization has the above-mentioned angle characteristics, and the emissivity in the direction of the incident angle θ _i is defined as ε (θ _i ), and normalization was performed for each incident angle.

And the intensity | strength I ( _ms , out) of the reflected light from the pipe inner surface part 12 in the final pipe | tube end part 11 to the exterior is represented by the following (8) Formula with the reflected light which calculated gradually and initial radiation. Use to calculate. In addition, “out” in the following equation (8) represents the direction in which light is emitted from the inner surface of the tube end to the outside. The initial value L ₀ (m, m _s ) was calculated from the surface emissivity calculated by experiment and its angular characteristics.

First, for a reflection parameter k = (K ₁ , K ₂ , σ) = (800, 200, 13.2) calculated from a general steel surface, the ratio L / d = 10 of the tube length L to the tube diameter d is set to FIG. 8 shows the acceptance angle characteristic of the apparent emissivity when the reflectance is changed. The vertical axis of the graph shown in FIG. 8 is the apparent emissivity, the horizontal axis is the light reception angle θ, and the light reception angle θ is 0 [°] for light reception in the direction orthogonal to the steel pipe axis direction. In the following, all wavelengths were set to 900 [nm].

  When the surface emissivity is high, it can be confirmed that measurement can be stably performed at an incident angle of 60 [°] or less. If it exceeds 60 [°], the emissivity decreases. FIG. 9 is a graph showing the results of evaluating the acceptance angle characteristics of the apparent emissivity when the emissivity is 0.9, separately for radiation and multiple reflection effects. From FIG. 9, it can be seen that the emissivity is reduced due to the radiation angle characteristic from the surface, not the multiple reflection effect. Therefore, it is desirable to measure so that the angle distribution of the radiated light intensity on the target surface becomes a light reception angle θ that can be regarded as constant.

Next, the presence / absence of specularity and the ratio L / d will be described. The surface having high specularity refers to a surface having a high specular reflection component coefficient K ₂ and a small σ indicating the spread of specular reflection compared to the diffuse reflection component coefficient K ₁ . Note that the larger σ is, the larger the specular reflection component is. FIG. 10 shows a graph when the ratio L / d is changed for a highly specular surface where the reflection parameter k = (K ₁ , K ₂ , σ) = (100, 1900, 0.57). FIG. 11 is a graph when the ratio L / d is changed for a surface with low specularity where the reflection parameter k = (K ₁ , K ₂ , σ) = (1900, 100, 17.2). Show. Note that the emissivity is 0.5 in both FIG. 10 and FIG.

  If the ratio L / d is equal to or greater than a certain value, the emissivity is stable except for the influence of the radiation angle characteristic of the tube inner surface portion 12. However, if the ratio L / d is small, the apparent angle is changed by the change in the light receiving angle θ. There are places where the emissivity of the abruptly changes. This characteristic is more remarkable as the specularity is higher. The mechanism will be described with reference to FIG. FIG. 12A is an explanatory diagram of the apparent emissivity when the tube inner surface portion exists in the regular reflection direction. FIG. 12B is an explanatory diagram of the apparent emissivity when the tube inner surface portion does not exist in the regular reflection direction. As shown in FIG. 12A, when the pipe inner surface portion 12 exists in the specular reflection direction, that is, in a direction symmetrical to the optical axis of the area sensor 2 with respect to the normal line of the tube end surface, the tube inner surface portion 12 is present. Is emitted from the tube inner surface portion 12 at the tube end portion 11 imaged by the area sensor 2 and received in addition to the light emitted from the tube inner surface portion 12 at the tube end portion 11. At this time, when the reflection property of the tube inner surface portion 12 has a high regular reflectance, the contribution of the added reflected light amount is large. On the other hand, as shown in FIG. 12B, when the tube inner surface portion 12 does not exist in the regular reflection direction, that is, in the direction symmetric to the optical axis of the area sensor 2 with respect to the normal line of the tube end surface, In addition to the radiated light from the tube inner surface portion 12 in the portion 11, the radiated light component is not received. Therefore, since the apparent emissivity varies greatly depending on the presence / absence of the tube inner surface portion 12 in the regular reflection direction, the ratio L / It is desirable to satisfy the conditions of d and the light receiving angle θ. Such conditions are shown in the following formula (1).

In addition, when the peeping angle is changed in FIG. 11, it does not change discretely depending on the presence or absence of the tube inner surface portion 12 in the regular reflection direction, but continuously changes over a certain region before and after that. This is because the specular reflection component reflected by the tube inner surface portion 12 at the tube end portion 11 and received by the area sensor 2 has not only the tube inner surface portion 12 where the incident angle and the reflection angle coincide with each other, but also the periphery thereof. It is. The spread of the specular reflection component depends on σ indicating the spread of the specular reflection of the reflection parameter k = (K ₁ , K ₂ , σ). The larger the σ is, the more the specular reflection component is in the tube inner surface portion 12 in the regular reflection direction. It includes a wide range of components in the vicinity, that is, changes gently with respect to the variation in the viewing angle.

  Therefore, as shown in the above equation (1), in consideration of the parameter of σ representing the spread of specular reflection among the reflection characteristics of the tube inner surface portion 12, among the light rays received by the area sensor 2, the specular reflection at the tube end surface. The tube inner surface portion 12 that contributes as light has a ratio L / d and a light receiving angle θ that includes a reflected light component that is within 2σ from the regular reflection condition, and is approximately 97.5 [%]. The above specular reflection component can be included. If the condition includes a reflected light component that is within 3σ from the regular reflection condition, it is possible to include a specular reflection component of 99.8 [%] or more, and measurement can be performed more stably.

  In addition, in FIG. 12, not only the first regular reflection direction but also the influence of the presence or absence of the tube inner surface portion 12 that is the position where the light is reflected and reflected twice is a slight variation factor of the apparent emissivity. Yes. Therefore, when the specularity of the tube inner surface portion 12 is extremely high, it is desirable that the specular reflection component reflected twice is sufficiently included in the light rays received by the area sensor 2. The conditions including the specular reflection component in that case are based on the same concept as the first specular reflection component, and if the following expression (9) is satisfied, it is possible to remove the specular reflection component of 97.5 [%] or more. Become. In addition, the following formula (9) assumes a Gaussian distribution with respect to the spread of the specular reflection, and the spread of the specular reflection in the case of reflecting twice can be derived from the convolution of the Gaussian distributions.

  In FIG. 11, when the ratio L / d changes to “2”, “3”, and “5”, the larger the ratio L / d is under the condition that the tube inner surface portion 12 is sufficiently present in the regular reflection direction. The reason why the emissivity is high is that the multiple reflection effect is reduced when the ratio L / d is too small. Therefore, it is better to avoid radiation temperature measurement under unstable conditions in which the apparent emissivity fluctuates greatly with respect to such a change in the light receiving angle θ.

Next, the case where a temperature difference arises in the pipe inner surface part 12 in the axial direction both ends of the steel pipe 10 is demonstrated. As shown in FIG. 13, the temperature of the tube end 11a on the side measured by the area sensor 2 is fixed at 1200 [° C.], and the temperature of the tube end 11b on the opposite side is 1000 [°]. It was set at three levels of [° C.], 1100 [° C.], and 1200 [° C.], and assumed a temperature distribution that was linear with respect to the position in the axial direction of the steel pipe. FIG. 14 shows the result of the measured temperature when the light receiving angle θ is changed when the temperature of the tube end 11b is 1000 [° C.], 1100 [° C.], and 1200 [° C.]. The ratio L / d is 10, the reflection condition of the tube inner surface portion 12 is k = (K ₁ , K ₂ , σ) = (800, 200, 13.2), the emissivity is 0.7, and the luminance changes from temperature to temperature. The apparent emissivity used in the conversion is 0.78.

  When the temperature of the tube end portion 11b is 1000 [° C.] and 1200 [° C.], the difference in measured temperature slightly increases from the time when the light receiving angle θ exceeds 60 [°], but the difference is 2 [° C.]. It was found that the influence of temperature measurement was small.

  From the above consideration, if the emissivity of the tube inner surface portion 12 at the tube end portion 11 is measured in advance, the apparent emissivity can be set and temperature can be measured in consideration of the influence of multiple reflection by simulation. Become. In order to further improve the accuracy of temperature measurement, it is desirable to devise the following two points.

  First, when the temperature is measured using the area sensor 2, there is a problem of camera shading. That is, when the area sensor 2 captures an image, the luminance at the edge of the field of view decreases due to the shading of the camera. Accordingly, the received light brightness varies depending on the visual field position even for the same object, and therefore, different temperatures are calculated. In general, the influence of shading increases as the angle becomes wider. Therefore, the first idea is to measure the received light luminance within the field of view at the same light amount with respect to the wavelength to be measured in advance, and when actually measuring the temperature, the received light luminance is calculated using the previously measured shading data. The temperature is calculated after correction. This makes it possible to obtain the same temperature as that at the center of the visual field even at the visual field end. The shading data can be obtained by acquiring radiance data of the same temperature at each visual field position and performing interpolation or curved surface approximation when performing temperature calibration using a black body furnace.

  Next, the second idea is a luminance calculation method. As for the luminance of the tube inner surface portion 12, an average value or the like may be obtained if the tube inner surface portion 12 is uniform. However, there are many cases where there is a portion where the luminance is reduced in the tube inner surface portion 12 due to abnormalities such as soot and oxide. Therefore, by acquiring the maximum value of the luminance of the tube inner surface portion 12 and converting it to temperature, the temperature can be calculated without being affected by soot and oxide. If there is noise in the area sensor 2 and the maximum value of the luminance of the tube inner surface portion 12 is not stable, it is possible to stabilize by averaging a plurality of luminances near the maximum value. For example, an average value of the luminances of five pixels from the tenth highest pixel to the sixth highest pixel may be taken.

  Further, in calculating the temperature of the tube inner surface portion 12, it is necessary to recognize the tube inner surface portion 12 from the obtained image. As a method for recognizing the tube inner surface portion 12, various methods such as general shape recognition such as template matching and Hough transform, and utilizing a luminance difference between the tube inner surface portion 12 and the tube outer surface portion 13 are conceivable.

  For example, when the temperature of the steel tube 10 is uniform, the tube inner surface portion 12 has a higher emissivity than the tube outer surface portion 13 due to the influence of multiple reflection, or when the temperature of the steel tube 10 being rolled is measured in an air atmosphere, In many cases, a difference in luminance due to a temperature difference or an apparent emissivity difference occurs between the tube inner surface portion 12 and the tube outer surface portion 13 such that the tube outer surface portion 13 is easier to cool than the tube inner surface portion 12. In this case, a histogram as shown in FIG. 15B is generated from an image of the pipe end portion 11 including the pipe inner surface portion 12 of the steel pipe 10 as shown in FIG. It is possible to stably extract the tube inner surface portion 12 by using the peak formed by the pixels of the tube outer surface portion 13 and the peak constituted by the pixels of the tube outer surface portion 13. In the case where the histogram has a high frequency component with respect to the luminance value and has a plurality of peaks other than the luminance of the tube inner surface portion 12 and the luminance of the tube outer surface portion 13, it is effective to remove the high frequency component by a low-pass filter. is there.

  Furthermore, after recognizing the tube inner surface portion 12, it is possible to estimate the light receiving angle θ by performing elliptic approximation and evaluating the ratio between the major axis and the minor axis. If the length of the major axis is a and the length of the minor axis is b, the light receiving angle θ can be calculated using the following equation (10). However, it is assumed that the cross-sectional shape in the direction orthogonal to the axial direction of the tube end face is a shape close to a circle.

  Then, the accuracy of temperature measurement can be further improved by correcting the apparent emissivity by using the information on the light receiving angle θ in correspondence with the simulation result and the experimental result. Specifically, the apparent emissivity is corrected by the following procedure.

  First, from the simulation results and the experimental results, the relationship between the light reception angle θ and the apparent emissivity is preliminarily formed into a table map. Next, the area sensor 2 acquires an image of the tube end portion 11 so as to include the tube inner surface portion 12. Next, the tube inner surface portion 12 is recognized from the obtained image by image processing by the image processing device 3. Next, after recognizing the tube inner surface portion 12, the arithmetic device 4 calculates the major axis length a and the minor axis length b from the tube end portion 11 by elliptic approximation. Next, the arithmetic unit 4 calculates the light reception angle θ using the calculated major axis length a and minor axis length b and the above equation (10). Finally, the arithmetic unit 4 determines the apparent emissivity from the relationship between the light reception angle θ and the apparent emissivity, which are previously table-mapped.

  The table map indicating the relationship between the light receiving angle θ and the apparent emissivity is updated by performing a simulation according to the pipe length L and the pipe diameter d of the steel pipe 10, the reflection parameter depending on the surface properties, and the like. By calculating the apparent emissivity using the light receiving angle θ calculated from the above, it is possible to further improve the calculation accuracy of the apparent emissivity, and thus improve the accuracy of temperature measurement.

  When the radiation temperature measurement of the steel pipe 10 is actually performed outside the heating furnace, the outermost pipe end portion in the steel pipe axial direction of the pipe inner surface portion 12 is deeper in the steel pipe axial direction than the outermost pipe end portion by natural cooling. There is a case that is cooled in comparison with the back portion of the tube. FIG. 16 shows a tube end image, a brightness profile in the longitudinal direction of the tube diameter center portion generated from the image, and an emissivity calculated by simulation for light reception angles θ of 45 [°] and 60 [°]. It is a thing.

  In the tube end portion image shown in FIG. 16, the position of the innermost tube end portion of the tube inner surface portion 12 and the position of the tube inner portion that is on the back side of the position of the outermost tube end portion are calculated by simulation. The change of the emissivity was very small and 0.02 or less for both the light receiving angle θ of 45 [°] and the light receiving angle θ of 60 [°]. This is about 3 [° C.] or less when converted into temperature assuming that the temperature of the tube inner surface portion 12 is 1200 [° C.] and the wavelength is 900 [nm]. On the other hand, as can be seen from the luminance profile shown in FIG. 16, the light receiving angle θ is 45 [°] and the light receiving angle θ is 60 [°] at the position of the tube end portion and the tube deep portion. It is considered that the temperature distribution is actually generated in the direction of the steel pipe axis of the pipe inner surface portion 12. Since the position of the outermost pipe end portion is often cut and not used as a product, the temperature measurement position required as the temperature of the pipe inner surface portion 12 is the position of the pipe inner portion in actual operation in making the material. . Further, when measuring the temperature of the tube inner surface portion 12, it is required that the tube inner surface portion 12 is not affected by the temperature drop degree of the tube end portion 11.

  FIG. 17 shows a tube end image, a luminance profile in the longitudinal direction of the tube diameter center generated from the image, and the presence or absence of a flat portion f having a substantially constant luminance and a flatness in the luminance profile. [°], 50 [°], 55 [°], and 60 [°] are shown. As can be seen from the tube end image shown in FIG. 17, the larger the light receiving angle θ, the more the image is taken by the area sensor 2 to the far side position in the steel tube axial direction of the tube inner surface portion 12. Further, comparing the light receiving angle θ of 45 [°] with the light receiving angle θ of 60 [°], first, in the luminance profile with the light receiving angle θ of 45 [°], the tube inner surface portion 12 starts from the tube end portion. The luminance continues to increase toward the innermost tube innermost portion which is the innermost side of the tube inner surface portion 12 in the end image, and there is almost no flat portion f where the luminance is substantially constant and leveled off. On the other hand, in the luminance profile with the light receiving angle θ of 60 °, the luminance reaches a constant luminance in the process from the most tube end portion of the tube inner surface portion 12 to the most tube deep portion, and there is a flat portion f.

  Here, when an image of the tube inner surface portion 12 is acquired and the temperature is calculated using the maximum luminance value or the average luminance in the vicinity of the maximum luminance value, the image of the tube end portion imaged by the change in the light receiving angle θ. The position of the innermost tube innermost portion 12 of the tube inner surface portion 12 changes. Therefore, when the innermost tube innermost portion 12 has a temperature gradient, the temperature measurement result changes due to a slight change in the light receiving angle θ.

  Since the light receiving angle θ changes depending on the alignment of the optical system and the position of the steel pipe 10, if the temperature measurement result changes depending on the light receiving angle θ, it becomes an error factor. Therefore, temperature measurement is performed under the condition that the innermost tube innermost portion 12 has as little temperature gradient as possible, that is, a constant luminance is reached in a process from the outermost tube end portion of the inner tube surface portion 12 to the innermost tube innermost portion and becomes a level state. . As a result, even if the light receiving angle θ is slightly changed, the influence on the measurement temperature is reduced, and it is possible to stably and practically measure the temperature of the tube inner surface portion 12 without being affected by the temperature drop of the tube end portion 11. .

  FIG. 18 shows that on the actual rolling line, the steel tube 10 having a wall thickness t of less than 30 [mm] or less, 30 [mm] or more and less than 40 [mm], or 40 [mm] or more has a light receiving angle θ of 45. It shows the brightness profile in the longitudinal direction of the tube diameter center portion generated from the tube end image captured by the area sensor 2 by changing [°], 50 [°], 55 [°], and 60 [°]. is there.

  The luminance profile shape with the light receiving angle θ of 50 [°], 55 [°], and 60 [°] is constant on the way from the most tube end portion of the tube inner surface portion 12 to the most tube inner portion at any wall thickness t. The brightness has been reached, and the slope of the graph is flat. On the other hand, the gradient profile of the luminance profile shape with the light receiving angle θ of 45 ° remains steep even when reaching the innermost tube innermost portion 12 at any wall thickness t. Therefore, it is possible to determine whether or not the temperature can be stably measured depending on whether the gradient of the graph corresponding to the tube inner surface portion 12 of the luminance profile shape is flat.

  The determination as to whether or not the flat portion f is present in the portion corresponding to the tube inner surface portion 12 of the luminance profile shape may be made visually from the tube end portion image. However, the index is calculated quantitatively by using signal processing. It is also possible to carry out. Various methods can be considered as the method of quantification. For example, it can be considered to evaluate the presence / absence of the flat portion f and the length of the flat portion f in the vicinity of the innermost tube innermost portion 12 of the tube inner surface portion 12. .

  Regarding the change in the tube diameter d of the steel pipe 10, if the tube diameter d is sufficiently large with respect to the wall thickness t of the steel pipe 10, the flat portion f is sufficiently present in the portion corresponding to the tube inner surface portion 12 of the luminance profile shape. Therefore, the influence on the light receiving angle θ is small. The reason will be briefly described. First, since the steel pipe 10 is very hot during drilling, the heat removal is dominated by radiation. Since the heat removal from the pipe outer surface portion 13 of the steel pipe 10 is uniform over the entire length of the steel pipe 10, it does not affect the temperature drop of the pipe end portion 11. Therefore, the radiation heat from the tube inner surface portion 12 toward the tube end is a main factor for the temperature drop of the tube end portion 11.

  Here, heat reception at the position of the innermost tube innermost portion 12 of the tube inner surface portion 12 in the visual field of the area sensor 2 is considered with a constant light receiving angle θ. First, the radiation heat from the tube inner surface portion 12 toward the tube end is constant because the solid angle of radiation to the tube end opening is equal regardless of the tube diameter d. Further, when the tube diameter d changes a times, the heat capacity (volume) and the surface area at the innermost tube innermost portion 12 are both a times, so the heat removal rate is not affected. Therefore, the change in the tube diameter d has little influence on the light receiving angle θ because the flat portion f is sufficiently present in the portion corresponding to the tube inner surface portion 12 of the luminance profile shape.

  First, observation is performed from the right end to the left end along the longitudinal direction of the tube diameter center portion of the tube end portion image shown in FIG. As can be seen from the luminance profile, the luminance is almost zero in the portion where the steel pipe 10 does not exist. Then, when the brightness reaches the tube end 11 from a state close to 0, the brightness starts to increase, and the brightness greatly increases from the most tube end portion of the tube inner surface portion 12 toward the deepest tube end portion. When the tube inner surface portion 12 reaches the tube outer surface portion 13 from the innermost tube deepest portion, the luminance rapidly decreases.

  Here, the ratio of the length of the flat portion f to the length of the tube inner surface portion 12 on the luminance profile is considered as an index Z as to whether or not the flat portion f exists in the portion corresponding to the tube inner surface portion 12 of the luminance profile shape. As a method for calculating the lengths of the tube inner surface portion 12 and the flat portion f, various methods such as straight line fitting are conceivable. Here, however, the luminance value is more than a certain ratio compared with the maximum luminance value of the luminance profile. Let the portion having a flat portion f. And if the flat part f exists in about a certain ratio or more of the pipe inner surface part 12, the flat part f exists sufficiently and temperature measurement can be performed sufficiently stably.

  Hereinafter, an example of an algorithm for calculating the length of the tube inner surface portion 12 and the length of the flat portion f in the luminance profile will be described with reference to FIG.

First, the maximum luminance value T _max of the luminance profile is calculated. Next, the luminance profile is searched from the right end to the left end. Specifically, a search is performed from the right end of the luminance profile in the left direction, and a portion exceeding the threshold value p1 × _Tmax is set as the start position P1 of the tube inner surface portion 12. A search is further made to the left from the start position P1 of the tube inner surface portion 12, and a portion exceeding the threshold value p2 × _Tmax is set as the start position P2 of the flat portion f. A search is further made to the left from the start position P2 of the flat portion f, and the portion below the threshold value p3 × _Tmax is set as the tube inner surface portion 12 and the end position P3 of the flat portion f. Then, the area between the start position P1 and the end position P3 in the luminance profile is defined as the tube inner surface portion 12, and the distance between the start position P1 and the end position P3 is calculated as the distance of the tube inner surface portion 12. Further, a region between the start position P2 and the end position P3 in the luminance profile is set as the flat part f, and the distance between the start position P2 and the end position P3 is calculated as the distance of the flat part f. Then, from the calculated distance between the tube inner surface portion 12 and the distance between the flat portions f, the ratio of the flat portion f in the tube inner surface portion 12 is calculated, and the ratio is used as an index Z.

For example, for each luminance profile shown in FIG. 18, the threshold p2 × T _max is set to 90 [%] (p2 = 0.9) of the maximum luminance value T _max and the start position P2 of the flat portion f is set. . The threshold value p1 × _Tmax sets the start position P1 of the tube inner surface portion 12 as a portion (p1 = 0.3) exceeding 30 [%] of the maximum luminance value _Tmax . Further, the threshold value p3 × _Tmax is a portion (p3 = 0.85) that is lower than 85% of the maximum luminance value _Tmax after the start position P2 of the flat portion f, and the tube inner surface portion 12 and the flat portion f. An end position P3 is set. Then, from the calculated distance between the tube inner surface portion 12 and the distance between the flat portion f, the proportion of the tube inner surface portion 12 occupied by the flat portion f is calculated as an index Z.

  In calculating the index Z, if the luminance profile shape is not gentle, a pre-processing for smoothing may be performed using an averaging filter, a median filter, or the like.

  FIG. 21 is a histogram of the index Z corresponding to each luminance profile in FIG. FIG. 22 is a graph showing the average and standard deviation of the index Z at each light receiving angle θ and each wall thickness t corresponding to each luminance profile in FIG.

  From the relationship between the light receiving angle θ and the index Z as shown in FIG. 21 and FIG. 22, if the light receiving angle θ is 50 [°] or more, more preferably 55 [°] or more, the flat portion f is stably formed. It becomes possible to measure the temperature. In addition, the larger the wall thickness t, the less the influence of cooling of the tube end 11 and the more stable temperature measurement at a position closer to the tube end. Therefore, when the wall thickness t is large, the light receiving angle of less than 50 [°]. In some cases, temperature measurement can be performed stably even with θ.

  The calculation method of the index Z described here is merely an example, and another index may be used as long as it can determine whether or not the flat portion f is sufficiently present. In addition, a threshold value may be provided for the index Z calculated in this method, and it may be determined whether or not the temperature can be stably measured. If the index Z is less than the threshold value, an operation that is incompatible may be performed. As the threshold value of the index Z, for example, when 0.4, as shown in FIG. 22, the light receiving angle θ is stable regardless of the wall thickness t at 50 [°], 55 [°], and 60 [°]. Temperature measurement.

[Example]
An embodiment of the steel pipe 10 of the present invention will be described. First, the surface emissivity of the steel material was measured. With respect to the surface temperature calculated in advance by embedding thermocouples at a plurality of depth positions near the surface in the same steel material as the steel pipe 10 and extrapolating using a one-dimensional heat transfer model (linear model), Using the calibrated area sensor 2, the tube end portion 11 was imaged and compared so as to include the tube inner surface portion 12. A wavelength of 900 [nm] was used. As a result of testing in the laboratory, the emissivity was found to be 0.7.

Next, parameters used for the simulation will be described. The pipe length L of the target steel pipe 10 was 8000 [mm], the pipe diameter d was 400 [mm], and the parameters of the surface properties were calculated as K ₁ = 800, K ₂ = 200, and σ = 13.2. Moreover, the radiation angle condition used the condition shown in FIG.

  FIG. 23 is a diagram illustrating a simulation result for each light receiving angle θ when the emissivity is 0.8. FIG. 24 is a diagram showing an arrangement position of the area sensor 2 with respect to the steel pipe 10. If the light receiving angle θ is 60 [°] or less, the emissivity can be set to 0.87 and stable measurement can be performed. Further, by setting the light receiving angle θ to 55 [°] or more and 60 [°] or less, the area sensor 2 can sufficiently look into the back side of the tube inner surface portion 12 while avoiding the influence of the temperature drop of the tube end portion 11. Is possible. Therefore, in this embodiment, the area sensor 2 is arranged so that the light receiving angle θ is 60 [°], and an image of the tube end portion 11 including the tube inner surface portion 12 of the steel tube 10 as shown in FIG. 25 is acquired. In this embodiment, the area sensor 2 is arranged so that the light reception angle θ is 60 [°], but the light reception angle θ is selected from the range of 55 [°] to 60 [°]. In addition, the area sensor 2 may be arranged to acquire the image. The area sensor 2 acquires temperature calibration and shading data in advance, recognizes the tube inner surface portion 12 from the obtained image, applies shading correction, and then calculates the maximum luminance. The method for recognizing the tube inner surface portion 12 may be implemented in terms of image processing. However, this time, because the temperature of the tube inner surface portion 12 is higher than that of the tube outer surface portion 13 due to multiple reflection and the influence of heat removal, the entire image is displayed. The maximum brightness was calculated. For stable measurement, a plurality of luminances near the maximum may be acquired and averaged instead of the maximum luminance.

Next, the third processing step in the temperature measurement method according to the present embodiment will be described. Here, a method for converting the obtained luminance into temperature will be described. A calibration method for a radiation thermometer such as the area sensor 2 is standardized by the method described in Non-Patent Document 3 above. First, the area sensor 2 is calibrated using a black body furnace and a standard radiation thermometer. That is, the temperature setting of the black body furnace is set at a plurality of points within the temperature region to be measured, and the indication value of the standard radiation thermometer and the luminance value (the amount of received light) of the area sensor 2 at that time are acquired. With respect to the plurality of temperature values T and luminance values V obtained, approximate curves are calculated using the following equation (11) to obtain parameters A, B, and C. Here, in the following equation (11), _{c 2} denotes a second constant radiation. Then, using the obtained parameters A, B, C and the luminance value V, the temperature is calculated by the following equation (12). In addition, since the following formula (12) is a calculation formula under the condition that the target is a black body, that is, the emissivity is 1, when the luminance value V is actually substituted, the value is divided by the apparent emissivity. There is a need.

The exposure time is set to 2000 [μsec], the camera gain is set to the same value as that at the time of calibration, and as a result of using a 12 [bit] camera as the area sensor 2, the parameters A, B, and C are respectively A = 8. .76 × 10 ⁻⁷ , B = 2.61 × 10 ⁻⁵ , C = 6.61 × 10 ⁷ , and as a result of calculating the temperature using the apparent emissivity 0.87 calculated by the above-mentioned simulation, The inner surface temperature of the end portion 11 was 1320 [° C.].

  Moreover, in the manufacturing process of the steel pipe 10 using the temperature information of the pipe inner surface part 12 measured using the temperature measuring method according to the present embodiment, the area sensor 2 is provided for the plurality of steel pipes 10 being conveyed on the manufacturing line. When the necessary luminance image is acquired, an image is always taken regardless of the presence or absence of the steel pipe 10 at the imaging position by the area sensor 2, and the pipe inner surface portion 12 is taken at the imaging position by the area sensor 2 by image processing. The necessary image may be selected so that the tube end portion 11 including the position is located and the capturing may be performed. In the present embodiment, temperature measurement is performed under an air atmosphere that is not affected by back light, and temperature measurement is performed in a state where there is no light emitter other than the steel pipe 10 to be measured in the visual field. Therefore, only when the average brightness within the image or within the specified range exceeds a preset threshold value, the image processing device 3 fetches the image at that time from the area sensor 2 and performs the above-described calculation. By performing the process of converting the luminance by the device 4 into a temperature, it is possible to measure the temperature associated with each steel pipe 10 on the production line.

  As described above, according to the temperature measurement method of the present invention, it is possible to model a complicated multiple reflection effect on the tube inner surface portion 12 at the tube end portion 11 of the steel tube 10 and accurately measure the temperature. It has become possible. The temperature measurement method of the present invention can be applied not only to the steel pipe 10 but also to the temperature measurement of the pipe inner surface portion at the pipe end portion of the polygonal pipe material.

DESCRIPTION OF SYMBOLS 1 Temperature measuring device 2 Area sensor 3 Image processing device 4 Arithmetic apparatus 10 Steel pipe 11 Pipe end part 12 Pipe inner surface part 13 Pipe outer surface part 20 Experimental apparatus 21 Heating furnace 22 Steel material sample 23 Brick 24 Thermocouple 25 Logger

Claims (6)

  The amount of light obtained from the inner surface of the tube end, which is the inner surface of the tube at the tube end of the tube-shaped object, is a self-luminous component directly emitted from the inner surface of the tube end, and the emitted light is emitted from the inner surface of the tube over the axial direction of the tube-shaped object. An apparent emissivity calculation method, wherein an apparent emissivity of the inner surface of the tube end is calculated as a sum of the multiple reflection components emitted from the inner surface of the tube end by multiple reflection. A temperature measuring method for measuring the temperature of the inner surface of the tube end that is the inner surface of the tube at the end of the tube-shaped object,
An image processing step for recognizing the tube end inner surface from an image of the tube end portion including the tube end inner surface;
The brightness value of the inner surface of the tube end recognized in the image processing step and the apparent emissivity of the inner surface of the tube end calculated by the apparent emissivity calculation method according to claim 1 are used. A temperature calculating step for calculating the temperature of the end inner surface;
A temperature measuring method characterized by comprising: A temperature measuring method for measuring the temperature of the inner surface of the tube end that is the inner surface of the tube at the end of the tube-shaped object,
The light receiving angle θ of the imaging means for imaging the tube end is 60 [°] or less,
When the inner diameter of the tube-shaped object is d, the length of the tube-shaped object in the axial direction is L, and σ is one of the parameters representing the surface properties of the inner surface of the tube, the following equation (1) is satisfied. A temperature measurement method characterized by satisfying.
In the temperature measurement method according to claim 3,
The light receiving angle θ is obtained by recognizing the tube end surface or the tube end inner surface at the tube end from the image of the tube end captured by the imaging unit, and the axial direction of the tube end surface or the tube end inner surface. A temperature measurement method comprising calculating using a cross-sectional shape in an orthogonal direction.   A tube material that is the tube-shaped object is manufactured using information on the temperature of the inner surface of the tube end measured using the temperature measurement method according to any one of claims 2 to 4. Production method. A temperature measuring device that measures the temperature of the inner surface of the tube end that is the inner surface of the tube at the tube end of the tube-shaped object,
Imaging means for imaging the tube end so as to include the tube end inner surface;
Image processing means for performing image processing for recognizing the inner surface of the tube end from the image captured by the imaging means;
Using the brightness value of the inner surface of the tube end recognized by the image processing and the apparent emissivity of the inner surface of the tube end calculated by the apparent emissivity calculation method according to claim 1, the tube end. Temperature calculating means for calculating the temperature of the inner surface;
A temperature measuring device comprising: