Temperature measuring device and temperature measuring method

Abstract

In a temperature measurement device for measuring the temperature of a steel plate, an imaging unit as a spectroscopic camera acquires the radiance of the steel plate at a plurality of wavelengths at each position of the surface of the steel plate within a field of view using a range including at least a part of the width direction of the steel plate as the field of view; the arithmetic processing unit (14) selects two wavelengths from the wavelengths at which the radiance of the steel plate has been acquired by the imaging unit, and calculates the temperature at each position on the surface of the steel plate by using the two-color radiation temperature measurement using the radiance of the steel plate acquired by the imaging unit at the two selected wavelengths.

Description

Technical Field

[0001] The technology disclosed herein relates to a temperature measurement device and a temperature measurement method, and more specifically, to a temperature measurement device and a temperature measurement method using radiance detected from an object. Background Art

[0002] Galvannealed steel sheets (hereinafter referred to as "GA steel sheets") are produced by heat-treating hot-dip galvanized steel sheets to form an iron-zinc alloy on the surface. They are excellent in corrosion resistance, paintability, and press workability, and are used, for example, in automobile exterior panels.

[0003] In the manufacturing process of GA steel sheet, there is an alloying process in which a zinc-iron plated alloy layer is formed on the surface of the steel sheet by immersing the steel sheet in molten zinc and then heating it in an alloying furnace. The iron content in the plated alloy layer during the alloying process is generally referred to as the alloying degree. Since the alloying degree has a great influence on the properties of GA steel sheet such as stamping formability and durability, it is important to control the alloying degree within an appropriate range.

[0004] Since the degree of alloying is related to the heating temperature and heating time during the alloying treatment, in order to produce a GA steel sheet with a uniform and high-quality plated alloy layer, it is important to control the temperature distribution in the width direction to be uniform and within an appropriate temperature range during the alloying process. Therefore, a technology for measuring the temperature distribution in the width direction is required during the alloying process.

[0005] As a remote temperature measurement method that can measure the temperature of an object at high speed without contact, there is radiation temperature measurement. Radiation temperature measurement is a method of measuring the temperature based on the intensity (radiation brightness) of the thermal radiation light emitted by an object corresponding to its temperature by a photodetector. In radiation temperature measurement, in order to accurately measure the temperature, it is necessary to understand the emissivity that varies according to the object and its surface state. However, in the manufacturing process of GA steel plates, during the process of alloying by the reaction of molten zinc and iron, the roughness of the steel plate surface changes, and the emissivity of the steel plate changes significantly from approximately 0.2 before alloying to approximately 0.8 after alloying. Therefore, it is difficult to use a general radiation thermometer to measure the surface temperature of the GA steel plate during the alloying process.

[0006] As a temperature measurement method using radiation thermometry, there is a two-color radiation thermometry. In the radiation brightness of the radiation light emitted from the object to be measured at temperature T [K], if two wavelengths λ A and B The emissivity is ε A and ε B , then at two wavelengths λ A and BThe radiance E in the case A and E B It can be expressed by equation (1) and equation (2). Here, L b (λ, T) is the radiation brightness at temperature T and wavelength λ in black body radiation that follows Planck's law, and can be expressed as in equation (3). In addition, c1 and c2 are Planck radiation constants.

[0007] [Formula 1]

[0008] E A =ε A L b (λ A ,T) (1)

[0009] E B =ε B L b (λ B ,T) (2)

[0010]

[0011] In the two-color radiation temperature measurement method, the temperature T is calculated by using the ratio of the radiation brightness of two wavelengths (two-color ratio R) as shown in formula (4). In particular, ε A =ε B The case is called gray body, and ε A With ε B By canceling out the influence of emissivity, the temperature T can be easily measured.

[0012] [Formula 2]

[0013]

[0014] Furthermore, the emissivity can be calculated from the calculated temperature T and the radiation brightness by using formula (1). Although the sensitivity is higher as the difference between the two wavelengths increases (the farther apart the wavelengths are from each other), it is generally difficult to satisfy the gray body condition in this case. In such a case, the temperature can be calculated by using some other method to obtain the emissivity and using the correct emissivity ratio.

[0015] As described above, in the GA steel sheet, if the steel sheet coated with molten zinc is heated to grow an alloy layer, the surface roughness of the steel sheet changes from a mirror surface state to a rough surface state as alloying progresses. In this case, the emissivity changes from about 0.2 (before alloying) to about 0.8 (after alloying). This corresponds to the fact that molten zinc has a metallic luster, and after alloying, it becomes a solid and a rough surface while generating FeZn compounds.

[0016] Wavelength λ of molten zinc and alloyed zinci Emissivity ε in [μm] i As shown in formula (5), the wavelength dependence formula shows that the wavelength dependence coefficient A1 = 0.1. Therefore, in the steel plate during the alloying process, generally at two wavelengths λ A and B In this case, the emissivity is not equal, but the relationship shown in equation (6), so the temperature cannot be calculated for the two-color radiation temperature measurement method that takes the gray body into consideration. Therefore, for example, as shown in Japanese Patent Laid-Open No. 2-85730 and Japanese Patent Laid-Open No. 5-231944, for a conventional radiation thermometer for GA steel plates in the alloying process, two or more wavelengths are used to measure the radiation brightness, and equations (1), (2), and (6) which is a relationship between the emissivity of the two wavelengths are combined and numerically analyzed to solve them, thereby calculating the temperature and the emissivity.

[0017] [Formula 3]

[0018]

[0019] The following method is proposed: using the above-mentioned temperature measurement method to measure the temperature distribution of an object with a variable emissivity. For example, Japanese Patent Publication No. 9-53992 discloses the following method: by scanning and splitting the radiated light with a rotating reflector, detecting the radiation brightness of multiple wavelengths in the scanning direction, and performing calculations based on the radiation brightness of the multiple wavelengths, the temperature distribution is measured.

[0020] In addition, as a method for measuring temperature distribution at high speed, there is a known method of measuring the distribution of thermal radiation light in space emitted from an object using a one-dimensional imaging element or a two-dimensional imaging element. In such a method, it is assumed that the emissivity is constant within the field of view, and as long as such a premise is established, temperature measurement without practical problems can be performed. For example, Japanese Patent Publication No. 2012-132442 discloses a method of measuring the distribution of emissivity and temperature by dividing the thermal radiation light emitted from the measurement object into three wavelengths and photographing it as a two-dimensional image of three wavelengths. Summary of the invention

[0021] Problems to be solved by the invention

[0022] However, the technique described in Japanese Patent Application Laid-Open No. 9-53992 detects the radiance while scanning with a rotating mirror. Therefore, the response speed depends on the rotation speed of the mirror, and there is a problem that it takes time to measure the width direction.

[0023] In addition, the technology described in Japanese Patent Publication No. 2012-132442 can measure the temperature distribution with high response because the spatial distribution of the radiation brightness is obtained by one imaging. However, the two-dimensional imaging element used in the imaging of the distribution image has a disadvantage of having a lower dynamic range than the photodetector used in the temperature measurement. For example, the radiation brightness is taken in as digital data of 1024 grayscale levels or 4096 grayscale levels. Therefore, if the detected radiation brightness is too high, it will be saturated, and if it is too weak, the brightness will be insufficient and the correct temperature cannot be measured. Therefore, it is necessary to adjust the sensitivity by adjusting the gain or exposure time of the imaging device in accordance with the temperature of the measured object so that the detection brightness level falls within the range in which the temperature measurement can be performed. However, in the case of a GA steel plate in the alloying process where the difference in emissivity is as much as 4 times, in the same field of view (image), parts with large emissivity and parts with small emissivity are mixed. In such a case, even if the sensitivity is adjusted, it is difficult to stably obtain the radiation brightness and correctly measure the temperature without excess or deficiency in both parts at the same time.

[0024] Therefore, an object of the technology disclosed in the present invention is to provide a temperature measurement device and a temperature measurement method, by which the temperature distribution of an object whose emissivity varies in a wide range can be measured with high response and high accuracy.

[0025] Means used to solve problems

[0026] The first technical solution of the technology disclosed in the present invention is a temperature measuring device for measuring the temperature of a steel plate, the temperature measuring device comprising: an imaging unit as a spectroscopic camera, which takes a range including at least a portion of the plate width direction of the above-mentioned steel plate as a field of view, and obtains the radiation brightness of the above-mentioned steel plate at multiple wavelengths at each position on the surface of the above-mentioned steel plate within the above-mentioned field of view; and an operation processing unit, which selects two wavelengths from the wavelengths of the radiation brightness of the above-mentioned steel plate obtained by the above-mentioned imaging unit, and uses the radiation brightness of the above-mentioned steel plate obtained by the above-mentioned imaging unit at the two selected wavelengths to calculate the temperature of each position on the surface of the above-mentioned steel plate by using two-color radiation temperature measurement.

[0027] According to a second technical solution of the technology disclosed in the present invention, in the temperature measuring device according to the first technical solution, the operation processing unit maintains at least two wavelength sets, which are composed of two wavelengths among the wavelengths of the radiation brightness of the steel plate obtained by the camera unit; when the position of the surface of the steel plate to be measured is in an area with large radiation brightness of the steel plate, two wavelengths included in a set of two wavelengths on the short wavelength side among the multiple wavelength sets maintained are selected; when the position of the surface of the steel plate to be measured is in an area with small radiation brightness of the steel plate, two wavelengths included in a set of two wavelengths on the long wavelength side among the multiple wavelength sets maintained are selected; and the radiation brightness of the steel plate obtained by the camera unit at the selected two wavelengths is used to calculate the temperature of each position on the surface of the steel plate by using two-color radiation temperature measurement.

[0028] According to a third technical solution of the technology disclosed in the present invention, in the temperature measuring device related to the second technical solution, the steel plate is an alloyed hot-dip galvanized steel plate; an area of ​​the steel plate with high radiation brightness corresponds to an area on the surface of the steel plate where alloying progresses smoothly; and an area of ​​the steel plate with low radiation brightness corresponds to an area on the surface of the steel plate where alloying does not progress.

[0029] According to a fourth technical solution of the technology disclosed in the present invention, in the temperature measuring device related to the second technical solution or the third technical solution, when the radiation brightness of the steel plate obtained by the wavelength set including the longest wavelength among the multiple wavelength sets maintained by the above-mentioned operation processing unit is lower than a preset lower limit threshold, the above-mentioned operation processing unit increases the exposure of the above-mentioned camera unit; when the radiation brightness of the steel plate obtained by the wavelength set including the shortest wavelength among the multiple wavelength sets maintained by the above-mentioned operation processing unit is higher than a preset upper limit threshold, the above-mentioned operation processing unit reduces the exposure of the above-mentioned camera unit.

[0030] According to a fifth technical solution of the technology disclosed in the present invention, in the temperature measuring device of any one of the technical solutions from the first technical solution to the fourth technical solution, the operation processing unit divides the field of view of the camera unit into a plurality of partitions along the width direction of the steel plate, and calculates the range of wavelengths at which the radiation brightness of the steel plate is successfully obtained by the camera unit for each of the partitions; selects two wavelengths from the range of wavelengths at which the radiation brightness of the steel plate is successfully obtained for each of the partitions; and uses the radiation brightness of the steel plate obtained by the camera unit at the two selected wavelengths to calculate the temperature of each position on the surface of the steel plate by utilizing two-color radiation temperature measurement.

[0031] According to a sixth aspect of the technology disclosed herein, in the temperature measuring device according to any one of the first to fourth aspects, the arithmetic processing unit selects, as the two wavelengths, wavelengths shorter than a wavelength at which the radiant brightness of the steel plate is maximum.

[0032] According to a seventh technical solution of the technology disclosed in the present invention, in a temperature measuring device according to any one of the first to sixth technical solutions, the operation processing unit detects the edge positions on both sides of the steel plate in the plate width direction according to the image obtained by the camera unit, and determines the range for measuring the temperature based on the detected edge positions.

[0033] The eighth technical solution of the technology disclosed in the present invention is a temperature measurement method for measuring the temperature of a steel plate, the method comprising: a brightness acquisition step, using a camera unit as a spectroscopic camera to acquire the radiation brightness of the steel plate at multiple wavelengths at each position on the surface of the steel plate within a field of view, the spectroscopic camera having a range including at least a portion of the plate width direction of the steel plate as a field of view; and a temperature calculation step, using a calculation processing unit to select two wavelengths from the wavelengths at which the radiation brightness of the steel plate is acquired by the camera unit, and using the radiation brightness of the steel plate acquired by the camera unit at the two selected wavelengths to calculate the temperature of each position on the surface of the steel plate by using two-color radiation temperature measurement.

[0034] According to a ninth technical solution of the technology disclosed in the present invention, in the temperature measurement method according to the eighth technical solution, the operation processing unit maintains at least two wavelength sets, which are composed of two wavelengths among the wavelengths of the radiation brightness of the steel plate obtained by the camera unit; in the temperature calculation step, when the position of the surface of the steel plate to be measured is in an area with large radiation brightness of the steel plate, two wavelengths included in a set of two wavelengths on the short wavelength side among the plurality of wavelength sets maintained are selected; when the position of the surface of the steel plate to be measured is in an area with small radiation brightness of the steel plate, two wavelengths included in a set of two wavelengths on the long wavelength side among the plurality of wavelength sets maintained are selected; and the radiation brightness of the steel plate obtained by the camera unit at the two selected wavelengths is used to calculate the temperature of each position on the surface of the steel plate by using two-color radiation temperature measurement.

[0035] According to a tenth technical solution of the technology disclosed in the present invention, in the temperature measurement method according to the ninth technical solution, the temperature measurement method is a method for measuring the temperature of the alloyed hot-dip galvanized steel plate as the steel plate; in the temperature calculation step, when the position of the surface of the steel plate to be measured is in an area where alloying of the surface of the steel plate progresses smoothly and the radiation brightness of the steel plate is large, two wavelengths included in a set of two wavelengths on the short wavelength side from among the plurality of wavelength sets retained are selected; when the position of the surface of the steel plate to be measured is in an area where alloying of the surface of the steel plate does not progress and the radiation brightness of the steel plate is small, two wavelengths included in a set of two wavelengths on the long wavelength side from among the plurality of wavelength sets retained are selected.

[0036] According to the eleventh technical solution of the technology disclosed in the present invention, in the temperature measurement method according to the ninth technical solution or the tenth technical solution, when the radiation brightness of the steel plate obtained by the wavelength set including the longest wavelength among the multiple wavelength sets maintained by the above-mentioned operation processing unit is lower than a preset lower limit threshold, the exposure amount of the above-mentioned camera unit is increased; when the radiation brightness of the steel plate obtained by the wavelength set including the shortest wavelength among the multiple wavelength sets maintained by the above-mentioned operation processing unit is higher than a preset upper limit threshold, the exposure amount of the above-mentioned camera unit is reduced.

[0037] According to a twelfth technical solution of the technology disclosed in the present invention, in the temperature measurement method of any one of the technical solutions from the eighth technical solution to the eleventh technical solution, in the temperature calculation step, the field of view of the camera unit is divided into a plurality of partitions along the width direction of the steel plate, and the range of wavelengths at which the radiation brightness of the steel plate is successfully obtained by the camera unit is obtained for each of the partitions; two wavelengths are selected from the range of wavelengths at which the radiation brightness of the steel plate is successfully obtained for each of the partitions; and the temperature of each position on the surface of the steel plate is calculated using two-color radiation temperature measurement using the radiation brightness of the steel plate obtained by the camera unit at the two selected wavelengths.

[0038] According to a thirteenth technical solution of the technology disclosed in the present invention, in the temperature measurement method according to any one of the eighth to twelfth technical solutions, in the temperature calculation step, wavelengths shorter than the wavelength at which the radiation brightness of the steel plate is maximum are selected as two wavelengths.

[0039] According to the fourteenth technical solution of the technology disclosed in the present invention, in the temperature measurement method of any one of the technical solutions from the eighth technical solution to the thirteenth technical solution, in the temperature calculation step, the edge positions on both sides of the steel plate in the plate width direction are detected according to the image obtained by the camera unit, and the range for measuring the temperature is determined based on the detected edge positions.

[0040] Effects of the Invention

[0041] According to the technology disclosed herein, it is possible to provide a temperature measurement device and a temperature measurement method that can measure the temperature distribution of an object whose emissivity varies in a wide range with high response and high accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a diagram showing the structure of a continuous hot-dip galvanizing treatment facility provided with the temperature measuring device according to the first and second embodiments.

[0043] Figure 2 This is a functional block diagram of the calculation processing unit of the temperature measurement device according to the first embodiment.

[0044] Figure 3 This is a block diagram showing the hardware configuration of the arithmetic processing unit.

[0045] Figure 4 This is a flowchart showing the flow of the temperature measurement process according to the first embodiment.

[0046] Figure 5 This is a functional block diagram of a calculation processing unit of the temperature measurement device according to the second embodiment.

[0047] Figure 6 This is a flowchart showing the flow of the temperature measurement process according to the second embodiment.

[0048] Figure 7 It is a graph showing the relationship between wavelength and brightness measured by a blackbody furnace.

[0049] Figure 8 It is a graph showing the relationship between the temperature of a blackbody furnace and the radiation brightness.

[0050] Fig. 9 It is a graph showing the relationship between the inverse of temperature and the radiance.

[0051] Fig.10 It is a figure which shows the structure of the test apparatus of Example.

[0052] Fig.11 It is a graph showing the temporal changes in brightness value and radiance.

[0053] Fig.12This is a graph showing an example of the calculation result of temperature.

[0054] Fig.13 It is a graph showing the temporal changes in brightness value and radiance.

[0055] Fig.14 This is a graph showing an example of the calculation result of temperature.

[0056] Fig.15 It is a diagram showing a measurement example of the temperature and the emissivity at each position in the width direction.

[0057] Fig.16 It is a schematic diagram showing the characteristics of a detector included in a spectroscopic camera. DETAILED DESCRIPTION

[0058] Hereinafter, an example of an embodiment of the technology disclosed herein will be described with reference to the drawings.

[0059] <First Embodiment>

[0060] (Description of the Structure of Temperature Measuring Device)

[0061] exist Figure 1 1 shows the structure of a continuous hot-dip galvanizing treatment equipment provided with a temperature measuring device 10 according to the first embodiment. After the plated steel sheet 1 is continuously annealed (not shown), it is introduced into a hot-dip galvanizing bath 2 filled with molten zinc. The steel sheet 1 introduced into the hot-dip galvanizing bath 2 is turned upward by the rollers 3 in the bath and is drawn out of the hot-dip galvanizing bath 2. For the steel sheet 1 with molten zinc attached, a wiping gas is sprayed from a wiping nozzle 4 toward both sides thereof, and the amount of molten zinc attached is adjusted. Then, the steel sheet 1 with the adjusted amount of plating attached is passed into an alloying furnace 5. The alloying furnace 5 heats the steel sheet 1 to an alloying temperature of about 500°C to 550°C by induction heating. The heated steel sheet 1 completes alloying before reaching the upper roller (not shown).

[0062] The temperature measurement device 10 according to the first embodiment is a device for measuring the temperature of a GA steel plate. Specifically, it is a device capable of measuring the temperature (temperature distribution) in a direction perpendicular to the conveying direction of the GA steel plate. The temperature measurement device 10 includes an imaging unit 12 and a calculation processing unit 14.

[0063] The camera unit 12 is composed of a hyperspectral camera, which includes a Figure 1The linear range of at least a portion of the direction orthogonal to the conveying direction of the GA steel plate shown (hereinafter also referred to as the "plate width direction") is the field of view, and the radiation brightness of each of the GA steel plate at multiple wavelengths is obtained at each position on the surface of the GA steel plate within the field of view. In addition, the "spectrographic camera" of the technology disclosed in the present invention is a camera that can split light and obtain information on each wavelength. With the conventional camera, i.e., the RBG camera, information related to the three wavelengths of light of red, blue, and green can be instantly obtained at each position photographed. However, if a spectroscopic camera is used, information related to, for example, more than 100 wavelengths of light can be instantly obtained at each position photographed. Spectroscopic cameras are sometimes also referred to as band hyperspectral cameras or multi-band cameras, etc., and the spectroscopic camera will be referred to as a hyperspectral camera hereinafter.

[0064] A hyperspectral camera that can be used as the imaging unit 12 is a camera that can capture spectral information obtained by spectrally dispersing light for each wavelength together with spatial information. Therefore, in order to obtain more information than that obtained by visual observation or RGB cameras, research is being conducted in the fields of engineering, biology, medicine, etc. Furthermore, in recent years, the equipment has become cheaper and is being introduced into factory production lines, and is also being used in the fields of foreign matter detection, quality management, component inspection, etc.

[0065] A general hyperspectral camera is a line scan type. It has a linear field of view, and by photographing the object, it can obtain subtle spectral information for each pixel that constitutes the line. In order to achieve this function, the hyperspectral camera is composed of a lens, a slit, a spectrometer, and a two-dimensional imaging element. First, the image of the object is formed on the slit through the lens, and the light formed on the slit reaches the spectrometer as one-dimensional information (line). In the spectrometer, the light is split into multiple wavelengths by a diffraction grating or a prism, and the split light is captured by a two-dimensional imaging element.

[0066] As a result, the two-dimensional imaging element of the hyperspectral camera obtains the detection intensity of light of each wavelength at each position in the linear field of view as a two-dimensional hyperspectral image of spatial CH (channel) × wavelength CH. As the spatial CH number, 320, 640, etc. are generally used, and as the wavelength CH number, 100, 200, etc. are used. The wavelength band of the spectroscopic spectrum is determined by the specifications of the spectrometer, the size of the two-dimensional imaging element, the detection band and the CH number. The detection intensity of light at each position is obtained in the hyperspectral image as a brightness value of 4096 gray levels (12 bits), for example. The hyperspectral camera of this embodiment with such a structure can also be said to have a structure with a one-dimensional diffraction grating.

[0067] If a hyperspectral camera having such characteristics is used, a spectral radiation spectrum of a wide range across the plate width direction of the GA steel plate can be acquired at high speed by a single imaging operation, so that a temperature measurement device 10 capable of high-speed temperature measurement can be realized.

[0068] In addition, by using one or appropriately multiple hyperspectral cameras, temperature measurement can be performed across the entire width of the GA steel plate in the plate width direction. Moreover, by sequentially measuring the GA steel plates passed through the plate in the rolling (length) direction, temperature measurement can be performed across the entire length of the GA steel plate. Therefore, temperature measurement across the entire length and width of the GA steel plate can also be performed.

[0069] Furthermore, by using two-color radiation temperature measurement, the two wavelengths used in the two-color radiation temperature measurement (i.e., a set of wavelengths consisting of two wavelengths) can be freely selected from the wavelengths of the spectral spectrum obtained by the hyperspectral camera, and changes in the emissivity of the object to be measured can also be tracked.

[0070] The imaging unit 12 captures one hyperspectral image for each line of view in the plate width direction of the GA steel plate. More specifically, the imaging unit 12 captures a hyperspectral image as an image containing radiation brightness of multiple wavelengths for each measurement point in the field of view for each line of view in the plate width direction of the GA steel plate.

[0071] In addition, the temperature of the GA steel plate in the alloying process is generally a temperature that can be regarded as following the Wien radiation law. Therefore, among the wavelengths of the radiation brightness obtained by the camera unit 12, a wavelength shorter than the wavelength at which the radiation brightness represented by the Wien displacement law at the temperature of the GA steel plate is maximum is selected.

[0072] The calculation processing unit 14 is a functional unit for calculating the temperature of the GA steel plate based on the hyperspectral image captured by the imaging unit 12, and is implemented by a hardware structure described below. The calculation processing unit 14 is functionally as follows: Figure 2 As shown, it includes a setting unit 26 , a selection unit 22 and a calculation unit 24 .

[0073] The setting unit 26 sets the exposure time of the imaging unit 12. Specifically, the setting unit 26 first conducts a test shot of the surface of the GA steel plate under a provisional condition by means of a hyperspectral camera provided in the imaging unit 12. For the hyperspectral image obtained by the test shot, when the radiant brightness of the GA steel plate obtained by a set of wavelengths including the longest wavelength among the multiple wavelength sets maintained by the selection unit 22 described later is lower than a preset lower limit threshold, the setting unit 26 sets the exposure time of the imaging unit 12 to a time longer than the currently set exposure time. Thus, the exposure amount of the imaging unit 12 is increased. Furthermore, when the radiant brightness of the GA steel plate obtained by a set of wavelengths including the shortest wavelength among the multiple wavelength sets maintained by the calculation processing unit 14 is higher than a preset upper limit threshold, the setting unit 26 sets the exposure time of the imaging unit 12 to a time shorter than the currently set exposure time. Thus, the exposure amount of the imaging unit 12 is reduced.

[0074] Next, the operation of re-imaging under the newly set provisional conditions and re-confirming the relationship between the radiant brightness and the threshold value is repeated until the imaging conditions of the imaging unit 12 become suitable for temperature measurement.

[0075] In this way, appropriate conditions can be determined as the imaging conditions of the imaging unit 12. As the hyperspectral image used in the following description, an image captured under the imaging conditions determined in this way can be used.

[0076] In addition, the setting unit 26 may set the exposure time of the imaging unit 12 to a time longer than the currently set exposure time, or alternatively, by setting the aperture value of the imaging unit 12 to a value larger than the currently set aperture value, thereby increasing the exposure amount of the imaging unit 12. In addition, the setting unit 26 may set the exposure time of the imaging unit 12 to a time shorter than the currently set exposure time, or alternatively, by setting the aperture value of the imaging unit 12 to a value smaller than the currently set aperture value, thereby reducing the exposure amount of the imaging unit 12.

[0077] If there is a margin in the imaging conditions of the imaging unit 12 or the temperature can be accurately calculated by the calculation unit 24 described later, the setting unit 26 may not be provided or the processing by the setting unit 26 may not be performed.

[0078] Here, the detector as a two-dimensional imaging element included in the hyperspectral camera can instantly obtain the two-dimensional brightness information (brightness value) incident from the GA steel plate, but Fig.16As shown in FIG. 1 , if the brightness value exceeds the appropriate detection range of the detector, the measured value becomes a saturated range, so the upper limit of the brightness value that can be detected is determined. Therefore, it is necessary to make adjustments to detect brightness values ​​that do not exceed the upper limit. In addition, if the brightness value is lower than the appropriate detection range of the detector, a deviation occurs in the measured value, so in order to accurately calculate the temperature, it is necessary to detect brightness values ​​above a certain level, and it is also necessary to determine the lower limit of the brightness value to be detected. Therefore, it is necessary to make adjustments to detect brightness values ​​that exceed the lower limit.

[0079] like Fig.16 Although there is an appropriate range where the actual radiance is roughly consistent with the measured radiance, the range is relatively narrow in the hyperspectral camera. Therefore, it is necessary to predetermine the appropriate range of the hyperspectral camera to be used. min And the upper limit of brightness I' max ) largely depends on the model of the hyperspectral camera, etc., so it is impossible to determine an appropriate range that is fixed in the temperature measurement device of this embodiment. The appropriate range is determined according to the model of the hyperspectral camera used, etc. In addition, the appropriate range is determined each time the model of the hyperspectral camera is changed. Moreover, the radiation brightness used in the calculation of the temperature of the GA steel plate (which will be described later, is the radiation brightness of two different wavelengths) is selected from the radiation brightness of multiple wavelengths obtained by the hyperspectral camera at each position photographed (i.e., the temperature measurement point in the plate width direction). A group of radiation brightness within the appropriate range is selected.

[0080] However, in the same hyperspectral image, when the emissivity varies greatly depending on the position, the upper and lower limits of the brightness value to be detected cannot be satisfied at the same time, and the correct temperature cannot be calculated across a large range in the plate width direction of the GA steel plate. For example, although the brightness value of the low emissivity portion can be properly detected, the brightness value of the high emissivity portion is saturated and cannot be detected, or although the brightness value of the high emissivity portion can be properly detected, the detection intensity of the low emissivity portion is insufficient and the brightness value cannot be accurately detected.

[0081] In addition, the radiation brightness (brightness value) is expressed as a function of temperature and wavelength. The wavelength at which the radiation brightness is maximum is determined by Wien's displacement law, and the wavelength [m] at temperature T [K] can be calculated as 0.0028978 / T. That is, for wavelengths shorter than it, the radiation brightness increases monotonically with the increase in wavelength. Therefore, in this embodiment, a hyperspectral image of a wavelength range in which the radiation brightness increases monotonically with respect to the wavelength can be obtained by a hyperspectral camera, and when the radiation brightness is extracted from the hyperspectral image, the two wavelengths used when obtaining the radiation brightness (i.e., a set of wavelengths consisting of two wavelengths) are switched according to the surface position of the GA steel plate as follows.

[0082] First, when the measurement point on the GA steel plate where the temperature is to be measured is included in a region with a large radiant brightness, the radiant brightness at a short wavelength is used, thereby reducing the radiant brightness compared to the long wavelength side, thereby preventing the influence of detector saturation. In addition, when the measurement point is included in a region with a small radiant brightness, the radiant brightness at a long wavelength is used, thereby increasing the radiant light intensity compared to the short wavelength side, thereby preventing the influence of insufficient radiant light intensity.

[0083] Therefore, the selection unit 22 holds at least two or more wavelength sets consisting of two wavelengths as the multiple wavelengths for obtaining the radiant brightness of the GA steel plate by the imaging unit 12. Furthermore, when the position of the surface of the GA steel plate whose temperature is to be measured is in an area where the radiant brightness of the GA steel plate is large, the selection unit 22 selects two wavelengths included in the set of two wavelengths on the short wavelength side from among the multiple wavelength sets held. On the other hand, when the position of the surface of the GA steel plate whose temperature is to be measured is in an area where the radiant brightness of the GA steel plate is small, the selection unit 22 selects two wavelengths included in the set of two wavelengths on the long wavelength side from among the multiple wavelength sets held.

[0084] The selection unit 22 compares the brightness with the preset appropriate brightness range to determine whether the position of the surface of the GA steel plate to be measured is in a region with a larger radiant brightness of the GA steel plate or in a region with a smaller radiant brightness. The appropriate brightness range is preset by the setting unit 26. That is, the setting unit 26 presets the brightness lower limit value I' of the appropriate brightness range. min And the upper limit of brightness I' max . Brightness lower limit I' min And the upper limit of brightness I' maxSince it depends on the performance of the hyperspectral camera used, it is set according to the model of the hyperspectral camera used. The selection unit 22 compares with the preset appropriate brightness range to determine whether the position of the surface of the GA steel plate to be measured is in a region where the radiant brightness of the GA steel plate is large or small. In addition, when it is determined that the position of the surface of the GA steel plate to be measured is in a region where the radiant brightness of the GA steel plate is large, the selection unit 22 selects two wavelengths included in a set of two wavelengths on the short wavelength side from the plurality of wavelength sets retained so that the radiant brightness of the GA steel plate obtained by the imaging unit 12 is within the preset appropriate brightness range. In addition, when it is determined that the position of the surface of the GA steel plate to be measured is in a region where the radiant brightness of the GA steel plate is small, the selection unit 22 selects two wavelengths included in a set of two wavelengths on the long wavelength side from the plurality of wavelength sets retained so that the radiant brightness of the GA steel plate obtained by the imaging unit 12 is within the preset appropriate brightness range. The selection unit 22 selects wavelengths shorter than the wavelengths at which the radiant brightness shown by the Wien displacement law at the temperature of the GA steel plate is maximum as the two wavelengths.

[0085] In addition, the area with a larger radiation brightness of the GA steel plate corresponds to the area where alloying of the surface of the GA steel plate progresses smoothly, and may be, for example, an area with an emissivity of about 0.1 to 0.3. In addition, the area with a smaller radiation brightness of the GA steel plate corresponds to the area where alloying of the surface of the GA steel plate does not progress, and may be, for example, an area with an emissivity of about 0.7 to 0.9.

[0086] In the case of a GA steel plate in the alloying process, the surface roughness changes during the alloying process, and the emissivity, which is approximately 0.2 before alloying, changes to approximately 0.8 after alloying. Therefore, even if the temperature is uniform, the radiation brightness of the portion where alloying has not progressed is small, and the radiation brightness of the portion where alloying has progressed smoothly becomes large, and the difference in the radiation brightness is as much as 4 times. Therefore, in the GA steel plate in the alloying process, the region where alloying has not progressed uses a set of wavelengths on the long wavelength side, and the region where alloying has progressed smoothly uses a set of wavelengths on the short wavelength side, so that there is no longer a situation where the required brightness in the temperature measurement cannot be obtained due to the small brightness in the region where alloying has not progressed, or the brightness is saturated due to excessive brightness in the region where alloying has progressed smoothly, and the brightness can be included in the dynamic range (appropriate range) of the camera unit, so the temperature distribution in the plate width direction can be accurately and stably measured.

[0087] In this way, the reference of the detection wavelength in the case of a measurement object having a large difference in emissivity depending on the position can be calculated as follows.

[0088] If the shortest wavelength of the detection wavelength under temperature T[K] is λ min , let the longest wavelength be λ max , then the emissivity change ratio (emissivity after change / emissivity before change, 4 in the case of GA steel plate) is equal to the radiation brightness ratio L b (λ max , T) / L b (λ min , T). In addition, L b (λ, T) is the radiant brightness of black body radiation at temperature T and wavelength λ. That is, if the emissivity change ratio is r e , by satisfying equations (7) and (8), the temperature can be stably measured even if the emissivity varies greatly depending on the position in the field of view (i.e., the position in the hyperspectral image). That is, it can be seen that a wavelength that satisfies this wavelength range can be selected using a hyperspectral camera.

[0089] [Formula 4]

[0090]

[0091] For example, the selection unit 22 holds a set on the short wavelength side including two wavelengths on the shorter wavelength side, and a set on the long wavelength side including two wavelengths on the longer wavelength side. In this case, if the first wavelength, the second wavelength, the third wavelength, and the fourth wavelength are sequentially set from the shorter wavelength side, the set on the short wavelength side is composed of the first wavelength and the second wavelength, and the set on the long wavelength side is composed of the third wavelength and the fourth wavelength. In this case, the second wavelength and the third wavelength may be the same or different.

[0092] In addition, as long as the main purpose of the technology disclosed in the present invention is followed, that is, by using the long wavelength side in the area where alloying has not progressed and using the short wavelength side in the area where alloying has progressed smoothly, it is possible to prevent the situation where the brightness required for temperature measurement cannot be obtained due to the low brightness in the area where alloying has not progressed, or the brightness is too strong and saturated in the area where alloying has progressed smoothly. The combination of wavelengths can be appropriately changed, such as using the first wavelength and the third wavelength to form a set on the short wavelength side and using the second wavelength and the fourth wavelength to form a set on the long wavelength side.

[0093] As described above, during the alloying process of the GA steel sheet, even at the same temperature, the radiant brightness increases monotonically with the increase in wavelength. In the imaging unit 12, saturation of radiant brightness is likely to occur at the fourth wavelength, and insufficient radiant brightness is likely to occur at the first wavelength. Therefore, even if the second wavelength is the same as the third wavelength or the combination of wavelengths is changed, there is a case where measurement can be performed, so the same wavelength or the combination of wavelengths can be used in this way.

[0094] The calculation unit 24 calculates the temperature of each position on the surface of the GA steel plate by two-color radiation temperature measurement using the radiance of the GA steel plate acquired by the imaging unit 12 at the two wavelengths selected by the selection unit 22 .

[0095] Here, in the case where the temperature of the GA steel plate to be measured varies greatly at the exit side of the heating furnace, etc., in the above-mentioned method (a method of comparing the radiant brightness of the GA steel plate with a preset appropriate brightness range to obtain the radiant brightness of two wavelengths selected so that the radiant brightness is within the preset appropriate brightness range), there is a case where the radiant brightness of two different wavelengths within the appropriate range (hereinafter also referred to as "a set of wavelengths") cannot be obtained in the range of simultaneous imaging (for example, the entire width of the steel plate at a specific length position in the length direction of the steel plate). For example, there is the following case: at a width position of a part of the width of the plate, when the temperature of the steel plate is very low or when the temperature of the steel plate is very high (or when alloying does not progress or when alloying progresses smoothly), the radiant brightness at this position is outside the appropriate range, and as a result, the temperature of the steel plate at this width position cannot be calculated. Even if a set of wavelengths within the appropriate range can be measured, there is a case where the two wavelengths have to be very close to each other, and the calculated temperature of the steel plate is incorrect. In such a case, it is necessary to follow the temperature distribution in the width direction of the steel plate to adjust the sensitivity of the imaging element used to detect the radiation brightness or the exposure time of the hyperspectral camera. As the sensitivity adjustment, it is possible to adjust the analog gain when reading the brightness value from the imaging element or control the exposure amount (at least one of the exposure time and the aperture).

[0096] For example, when measuring the temperature of 400°C to 600°C by detecting the radiant brightness at a near-infrared wavelength of 1500, the radiant brightness doubles for every 50°C rise in temperature. Therefore, in order to ensure a measurement range of 200°C, it is necessary that the measured radiant brightness is within an appropriate range even if the radiant brightness changes 16 times. In this case, if it is assumed that the minimum brightness required to correctly calculate the temperature is about 500, the imaging element must not be saturated until the radiant brightness reaches at least 8000. However, the dynamic range (appropriate range of radiant brightness) of a general two-dimensional imaging element is narrow. For example, in the case of 12 bits, it saturates at approximately 4000 as the upper limit of the brightness. Therefore, in actual measurement, it is desirable to control the sensitivity of the imaging element (for example, change the exposure time) to ensure the detection intensity that can accurately calculate the temperature.

[0097] At this time, no matter where in the width direction of the plate to be measured, at least one set of wavelength sets must be able to detect the radiation brightness at that position, so when all wavelength sets are reviewed, it is necessary to satisfy the condition that the minimum brightness to the maximum brightness of the GA steel plate can be detected by a certain wavelength included in all sets (hereinafter simply referred to as "condition"). Specifically, a spectroscopic camera is used in which the ratio of the radiation brightness detected at the shortest wavelength to the radiation brightness detected at the longest wavelength at the measured temperature is larger than the maximum change ratio of the emissivity (0.8 / 0.2=4 in the case of the GA steel plate), and the radiation brightness in the case of the wavelength set on the long wavelength side (the wavelength on the short wavelength side included therein) is controlled to be above the minimum brightness, and the radiation brightness in the case of the wavelength set on the short wavelength side (the wavelength on the long wavelength side included therein) is controlled to not exceed the upper limit of the brightness. By doing so, the wavelength set on the long wavelength side can be used when the radiation brightness in the case of the wavelength set on the long wavelength side is not saturated, and the wavelength set on the short wavelength side can be used when the radiation brightness in the case of the wavelength set on the long wavelength side is saturated, so as to perform temperature calculation. In addition, the conditions described here only need to be satisfied at the time when the entire wavelength set is reviewed, and may not be satisfied at times before or after this time.

[0098] In addition, the so-called two-color radiation temperature measurement generally refers to a method of calculating the temperature based on the radiation brightness obtained under the two wavelengths when the ratio of the emissivity under the two wavelengths (emissivity ratio) is fixed or can be regarded as fixed. However, in the technology disclosed in the present invention, not only such a case, but also a case where the emissivity ratio is not constant (depending on the progress of the reaction, etc.) but is variable is also called two-color radiation temperature measurement. In the temperature measurement related to the present embodiment, it is desirable to investigate in advance what value the emissivity ratio takes according to the wavelength used (the set of wavelengths used).

[0099] Figure 3 1 is a block diagram showing the hardware structure of the operation processing unit 14. Figure 3 As shown, the arithmetic processing unit 14 includes a CPU (Central Processing Unit) 32, a RAM (Random Access Memory) 34, a storage device 36, an input device 38, an output device 40, a storage medium reading device 42, a communication I / F (Interface) 44, and a bus 46. The bus 46 is a signal line that connects the CPU 32, the RAM 34, the storage device 36, the input device 38, the output device 40, the storage medium reading device 42, and the communication I / F 44 so that they can communicate with each other.

[0100] The storage device 36 stores a program for executing the temperature measurement process described later. The CPU 32 is a central processing unit that executes various programs or controls various components. That is, the CPU 32 reads the program from the storage device 36 and executes the program using the RAM 34 as a work area. The CPU 32 controls the above-mentioned components and performs various calculations according to the program stored in the storage device 36.

[0101] The RAM 34 temporarily stores programs and data as a work area. The storage device 36 is composed of a ROM (Read Only Memory), a HDD (Hard Disk Drive) or an SSD (Solid State Drive), etc., and stores various programs including an operating system and various data.

[0102] The input device 38 is a device for performing various inputs, such as a keyboard and a mouse. The output device 40 is a device for outputting various information, such as a display and a printer. A touch panel display may be used as the output device 40, thereby also functioning as the input device 38.

[0103] The storage medium reading device 42 reads data stored in various storage media such as CD (Compact Disc)-ROM, DVD (Digital Versatile Disc)-ROM, Blu-ray Disc, USB (Universal Serial Bus) memory, etc., and writes data to the storage medium. The communication I / F 44 is an interface for communicating with other devices by wire or wirelessly, for example, using standards such as Ethernet (registered trademark), FDDI or Wi-Fi (registered trademark).

[0104] The computer is composed of CPU 32, RAM 34 and storage device 36. The computer can be configured as a subcomputer that controls the action of a part of the temperature measuring device 10, or as a main computer that controls the overall action of the temperature measuring device 10. In part or all of the computer, for example, an integrated circuit such as LSI (Large Scale Integration) or an IC (Integrated Circuit) chip set can also be used. In the integration of the computer, it is not limited to LSI, and a dedicated circuit or processor can also be used. In addition, the processor described here is a processor in a broad sense, which can be a general-purpose processor or a dedicated processor including GPU (Graphics Processing Unit), ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array). In addition, the action of the processor can also be completed not by only one processor, but by the cooperation of multiple processors physically present in separate locations.

[0105] (Explanation of temperature measurement method)

[0106] Figure 4 14 is a flowchart showing the flow of the temperature measurement process executed by the CPU 32 of the calculation processing unit 14. The CPU 32 reads a program for executing the temperature measurement process from the storage device 36, expands it in the RAM 34 and executes it. As a result, the CPU 32 functions as each functional unit of the calculation processing unit 14 and executes Figure 4 The temperature measurement process shown is as follows. Thus, the temperature measurement method is executed in the temperature measurement device 10.

[0107] In step S10, the selection unit 22 obtains the hyperspectral image captured by the imaging unit 12. Next, in step S12, the selection unit 22 determines whether the position (measurement point) on the GA steel plate to be measured for temperature is in a region with a large radiation brightness or in a region with a small radiation brightness, and proceeds to step S14. When making this determination, it is possible to use an arbitrary wavelength λ preset at the measurement point based on the hyperspectral image. i The brightness of the radiation in the case is used as the basis for judgment. For example, the wavelength λ at the measurement point can be iWhen the radiation brightness is above a preset threshold, the measurement point is determined to be in an area with a larger radiation brightness. When it is below a preset (other) threshold, the measurement point is determined to be in an area with a smaller radiation brightness. As a threshold, for example, it is sufficient to predetermine a threshold such as a measurement point that is determined to be in an area with a larger radiation brightness when the emissivity is 0.1 to 0.3 (i.e., an area where alloying has not progressed) and a measurement point that is determined to be in an area with a smaller radiation brightness when the emissivity is 0.7 to 0.9 (i.e., an area where alloying has progressed smoothly). As a threshold other than this, for example, a predetermined appropriate range of radiation brightness (specifically, a lower brightness limit value I') may be used according to the model of the hyperspectral camera, etc. min And the upper limit of brightness I' max ).

[0108] In step S14, (i) in step S12, when it is determined that the measurement point is in an area with a large radiant brightness, a set on the short wavelength side (that is, two wavelengths of short wavelengths) is selected as two wavelengths used in temperature calculation from the plurality of wavelength sets held by the selection unit 22. On the other hand, (ii) in step S12, when it is determined that the measurement point is in an area with a small radiant brightness, a set on the long wavelength side (that is, two wavelengths of long wavelengths) is selected as two wavelengths used in temperature calculation from the plurality of wavelength sets held by the selection unit 22.

[0109] Next, in step S16, the calculation unit 24 uses the radiation brightness at the two wavelengths selected in step S14 to calculate the temperature of the measurement point on the surface of the GA steel plate by two-color radiation temperature measurement. In addition, the radiation brightness (two) at the two wavelengths selected in step S14 can be obtained from the brightness value of the hyperspectral image captured by the imaging unit 12.

[0110] Next, in step S18, the calculation unit 24 determines whether to end the processing of the hyperspectral image that is currently being processed. For example, since the processing is performed for each measurement point in steps S12 to S16, in step S18, when it is detected that the measurement point reaches the end of the hyperspectral camera, etc., with respect to the hyperspectral image acquired in step S10, it is determined that the processing of all measurement points in the hyperspectral image is completed, and the process proceeds to step S20. On the other hand, when it is determined that the processing of all measurement points is not completed and the processing of the hyperspectral image that is currently being processed cannot be completed, the process returns to step S12, and the measurement point is changed to perform the processing from step S14 onwards.

[0111] Next, in step S20, it is determined whether all the hyperspectral images to be measured have been processed. If not, the process returns to step S10, changes the camera position, etc., obtains the hyperspectral image again, and performs the processes below step S12. On the other hand, in step S20, if it is determined that all the hyperspectral images to be measured have been processed, the temperature measurement process ends. Step S10 is an example of a "brightness acquisition step" of the technology disclosed in the present invention, and steps S12 to S16 are examples of a "temperature calculation step" of the technology disclosed in the present invention.

[0112] (Effect of temperature measuring device)

[0113] As described above, the temperature measuring device according to the first embodiment changes the combination of two wavelengths according to the radiation brightness of the GA steel plate, and calculates the temperature of each position on the surface of the GA steel plate by using the two-color radiation temperature measurement method using the two wavelengths. Specifically, for the measurement point in the area with greater radiation brightness, two wavelengths on the short wavelength side are selected from a plurality of wavelengths, and for the measurement point in the area with less radiation brightness, two wavelengths on the long wavelength side are selected from a plurality of wavelengths to calculate the temperature. Thus, the temperature distribution of an object whose emissivity changes in a wide range can be measured with high response and high accuracy.

[0114] <Second Embodiment>

[0115] (Description of the Structure of Temperature Measuring Device)

[0116] In the temperature measurement device according to the second embodiment, the same components as those of the temperature measurement device 10 according to the first embodiment are denoted by the same reference numerals, and detailed description thereof will be omitted.

[0117] like Figure 1 As shown, the temperature measurement device 210 according to the second embodiment includes an imaging unit 12 and a calculation processing unit 214 .

[0118] The calculation processing unit 214 is a functional unit for calculating the temperature of the GA steel plate based on the hyperspectral image captured by the imaging unit 12, and has the same hardware configuration as that of the first embodiment ( Figure 3 ) is implemented. The operation processing unit 214 is functionally as follows Figure 5 As shown, it includes a selection unit 222 and a calculation unit 224.

[0119] The selection unit 222 divides the field of view of the imaging unit 12 (i.e., the range of the hyperspectral image) into a plurality of subareas along the plate width direction of the GA steel plate, and obtains the range of wavelengths in which the imaging unit 12 successfully obtains the radiant brightness of the GA steel plate for each subarea. Furthermore, the selection unit 222 selects two wavelengths from the range of wavelengths in which the radiant brightness is successfully obtained (i.e., the saturation of the brightness value or the lack of brightness value) for each subarea.

[0120] In addition, the partition can be determined with any length, but it is preferably determined in a manner that the alloying degree of the GA steel sheet within the range of one partition does not increase much, that is, the radiant brightness does not change much. For example, there is no need to specifically limit the distribution of temperature measurement points in the plate width direction (width direction resolution), but it can be set to 5 mm or less, for example. In addition, there is no need to specifically limit the measurement length in the conveying direction of the GA steel sheet, and it can be set to approximately 5 mm, for example.

[0121] In addition, the range in the plate width direction of the GA steel plate in multiple partitions (i.e., the range in which the temperature is measured) is preferably the full range in the plate width direction. The full range in the plate width direction can be detected by first acquiring a hyperspectral image, detecting the edge positions on both sides of the GA steel plate in the plate width direction based on the hyperspectral image, and grasping the range of the GA steel plate on the hyperspectral image based on the detected edge positions. As a method for detecting the edge position, a method such as differentiating the plate width distribution of the radiant brightness to explore its maximum and minimum values ​​can be used. In addition, the method for determining the two edge positions in the plate width direction in the present embodiment can also be applied to other embodiments.

[0122] As described above, the two-dimensional imaging element or detector included in the hyperspectral camera can instantaneously obtain the two-dimensional brightness information (brightness value) incident from the GA steel plate, but if the brightness value exceeds the appropriate detection range of the detector, the measured value becomes a saturated range, so the upper limit of the brightness value that can be detected is determined. Therefore, it is necessary to adjust to detect brightness values ​​that do not exceed its upper limit. In addition, in order to accurately calculate the temperature, it is necessary to detect brightness values ​​above a certain level, and it is also necessary to determine the lower limit of the brightness value to be detected. Therefore, it is necessary to adjust to detect brightness values ​​that exceed its lower limit. If the observation method is changed, in the case where the adjustment is insufficient and the brightness value of a certain measurement point exceeds the upper or lower limit of the detector and becomes a measurement error, it can also be said that the wavelength that measures such a brightness value cannot be used for temperature measurement at the measurement point. Conversely, it can also be said that as long as it is a wavelength that can obtain certain measurement data, it can be used for temperature measurement without exceeding the upper or lower limit of the detector.

[0123] Therefore, in this embodiment, the feature of being able to obtain a spectroscopic spectrum in a wide band of the hyperspectral camera is effectively utilized, and the field of view of the hyperspectral camera is divided into a plurality of smaller sub-areas along its width direction, which can be regarded as having almost no difference in the degree of alloying, and two wavelengths are selected from wavelengths where the brightness is successfully obtained (i.e., not exceeding the upper limit of the detector and not falling below the lower limit) for each of the sub-areas. In this way, stable temperature measurement with good accuracy can be performed.

[0124] The calculation unit 224 calculates the temperature of each position on the surface of the GA steel plate by two-color radiation temperature measurement using the radiance of the GA steel plate acquired by the imaging unit 12 at the two wavelengths selected by the selection unit 222 for each section.

[0125] (Explanation of temperature measurement method)

[0126] Figure 6 2 is a flowchart showing the flow of temperature measurement processing executed by the CPU 32 of the calculation processing unit 214. The CPU 32 reads a program for executing the temperature measurement processing from the storage device 36, expands it into the RAM 34 and executes it, thereby the CPU 32 functions as each functional unit of the calculation processing unit 214 and executes Figure 6 Thus, the temperature measurement method is executed in the temperature measurement device 210 .

[0127] In step S10, the selection unit 222 obtains the hyperspectral image captured by the imaging unit 12. Next, in step S210, the selection unit 222 determines the range of wavelengths in which the radiance is successfully obtained for each of the plurality of sub-areas in the field of view of the imaging unit 12. More specifically, the selection unit 222 determines the range in which the radiance is successfully obtained for the sub-area including the position (measurement point) at which the temperature is to be measured.

[0128] Next, in step S212, the selection unit 222 selects two wavelengths λ from the determined wavelength range for each partition. A and B Next, in step S214, the calculation unit 224 uses the selected two wavelengths λ for each partition. A and B The radiance E in the case A and E B , the two-color radiation temperature measurement method is used to calculate the surface temperature of the GA steel plate.

[0129] Next, in step S216, the calculation unit 224 determines whether to end the processing of the hyperspectral image currently being processed. If the processing is ended, the process proceeds to step S218, and if the processing is not ended, the process returns to step S210, changes the measurement point, and performs the processing below step S210.

[0130] Next, in step S218, it is determined whether all the hyperspectral images to be measured have been processed. If not, the process returns to step S10, changes the camera position, etc., obtains the hyperspectral image again, and performs the processes below step S12. On the other hand, in step S218, if it is determined that all the hyperspectral images to be measured have been processed, the temperature measurement process is terminated. Steps S210 to S214 are an example of a "temperature calculation step" related to the technology disclosed herein.

[0131] (Effect of temperature measuring device)

[0132] As described above, the temperature measurement device according to the second embodiment divides the field of view of the imaging unit into a plurality of partitions along the plate width direction of the GA steel plate, and finds the wavelength range in which the imaging unit successfully obtains the radiant brightness of the GA steel plate for each partition. Furthermore, for each partition, two wavelengths are selected from the wavelength range in which the radiant brightness is successfully obtained, and the temperature is calculated. Thus, the temperature distribution of an object whose emissivity varies in a wide range can be measured with high response and high accuracy.

[0133] (Regarding a more preferred embodiment)

[0134] As described in detail above, when using a hyperspectral camera to measure the temperature distribution in the width direction of a GA steel plate, there occurs a problem that although the brightness value of the low emissivity portion can be properly detected, the brightness value of the high emissivity portion is saturated and cannot be detected; and a problem that although the brightness value of the high emissivity portion can be properly detected, the detection intensity of the low emissivity portion is insufficient and the brightness value cannot be accurately detected. In order to solve this problem, in a more preferred form, for the plate width direction of the GA steel plate, the range of measuring temperature distribution is determined, the field of view of the camera unit 12 is divided into a plurality of partitions along the plate width direction of the GA steel plate, an appropriate brightness range is set according to the performance of the hyperspectral camera used, the exposure is adjusted so that the radiation brightness of the GA steel plate obtained by the camera unit 12 is included in the appropriate brightness range, and the range of wavelengths in which the camera unit 12 successfully obtains the radiation brightness of the GA steel plate is obtained for each partition. For each partition, two wavelengths are selected from the range of wavelengths in which the radiation brightness is successfully obtained (i.e., no saturation of the brightness value or insufficient brightness value occurs). For each partition, the radiation brightness of the GA steel plate obtained by the camera unit 12 at the selected two wavelengths is used to calculate the temperature of each position on the surface of the GA steel plate using two-color radiation temperature measurement.

[0135] <Example>

[0136] An example will be described using a GA steel sheet in the alloying process as a subject.

[0137] (Specifications required for the sensor and the hyperspectral camera used)

[0138] Table 1 shows the required specifications of the temperature measuring device. The temperature of the GA steel plate in the alloying process is 400°C to 600°C, and the emissivity changes from approximately 0.2 (before alloying) to approximately 0.8 (after alloying) due to alloying. This is because molten zinc exhibits a metallic luster, while, in contrast, after alloying, it becomes a solid and becomes a rough surface as the FeZn compound is generated. The distribution of temperature measurement points in the plate width direction (resolution in the width direction) is set to less than 5 mm, the temperature measurement cycle in the plate width direction (response speed) is set to 20 Hz, and the measurement accuracy target is set to a difference of ±10°C from the actual measured value, so that the relationship between the various parts of the steel plate that has been passed through the plate and alloyed and the coating quality can be grasped.

[0139] [Table 1]

[0140]

[0141] Table 2 shows the specifications of the hyperspectral camera used as the imaging unit 12. As a two-dimensional imaging element, in order to ensure the sensitivity of being able to fully detect the radiated light even when the object temperature is 400°C and the emissivity is 0.2, a hyperspectral camera using an InGaAs light detection element with sensitivity from 900nm to 1650nm is selected. The number of CHs in the spatial direction is 640CH, and a lens with an image angle of 65° in the spatial direction is used. If the distance from the measurement point is 1670mm, the field of view in the board width direction is 2000mm, and the resolution in the board width direction is approximately 3mm. On the other hand, the wavelength CH is 224CH, and for the range from 900nm to 1700nm, the spectral spectrum can be obtained at a pitch of 3.5nm. The maximum frame rate is 50fps, and the sensitivity can be controlled by setting the exposure time.

[0142] [Table 2]

[0143]

[0144] The hyperspectral image output from the hyperspectral camera is output as an image of 640 pixels in horizontal direction (spatial direction CH) × 224 pixels in vertical direction (wavelength CH) and 4096 grayscale levels (12 bits), and is taken into the calculation processing unit 14. In the calculation processing unit 14, the radiance E of multiple wavelengths is calculated. i , used for temperature measurement.

[0145] In the following embodiment, an example of selecting two wavelengths from four wavelengths obtained by a hyperspectral camera to perform temperature measurement is shown, but in terms of expanding the temperature measurement range, the number of wavelengths is not limited to 4. As long as the number of wavelengths is effective within the wavelength range of the hyperspectral camera, more wavelengths such as six wavelengths may be selected.

[0146] (Choice of wavelength)

[0147] exist Figure 7 In the figure, for the GA steel plate measurement temperature range of approximately 400°C, 500°C and 600°C, the spectral spectra (the brightness of the blackbody furnace at each wavelength) when the blackbody radiation in the high temperature part of the blackbody furnace is measured by a hyperspectral camera are shown. Figure 7 The value corrected for the background noise (background) (the detection brightness value in a state where no light enters at all, generated by the dark current of the two-dimensional imaging element) is shown in . At any temperature, the exposure time of the blackbody furnace is set in such a way that the radiation light intensity (blackbody furnace measurement brightness) becomes approximately the same. A temperature rise of about 100°C is equivalent to 1 / 4 of the exposure time. The radiation light intensity increases toward the longer wavelength side, and is maximum at around 1600nm according to the sensitivity characteristics of the detection element (InGaAs) used. The detection wavelengths are set to 4: λ1=1300nm, λ2=1400nm, λ3=1500nm, and λ4=1600nm. In the two-color radiation temperature measurement method, if the wavelength interval is narrow, the sensitivity decreases. In order to ensure sufficient sensitivity to meet the required accuracy this time, a wavelength interval of 100nm is set. The alloying temperature is around 500°C, L b (1600nm, 500℃) / L b (1300nm, 500℃) = 5.18. This value is larger than the change ratio of the emissivity before and after alloying, which is 4 (= 0.8 / 0.2). Even if there is uneven alloying in the field of view, temperature measurement can be performed by using a certain wavelength. In addition, when extracting brightness values ​​(radiant light intensity) from hyperspectral images, in order to stabilize the brightness values, the average value of multiple CHs (e.g. 5CH) can also be used in the wavelength direction.

[0148] (Temperature correction in blackbody furnace)

[0149] exist Figure 8 The results are shown in Figure 1. The black body furnace (T[℃] = 300-600) is photographed at each set exposure time, the radiation brightness of each wavelength is extracted from the hyperspectral image, and the relationship with temperature is plotted as a curve. The set exposure time t d [μs] is t d [μs] = 1000, 1600, 2560, 4096, 6553, 10485, 16777, 26489, the detection wavelengths are λ1 = 1300nm, λ2 = 1400nm, λ3 = 1500nm, λ4 = 1600nm. Figure 8The value after subtracting the background noise is shown in . Since the background noise varies depending on the exposure time, the shutter is closed for each exposure time so that light does not enter the two-dimensional imaging element and the measurement is performed. If the exposure time is different, the sensitivity of the hyperspectral camera is different, so the correlation with temperature is no longer obtained. Therefore, the brightness value I extracted from the hyperspectral image is obtained using formula (9) i With the reference exposure time t dst = 4096μs is converted to the radiant brightness E i I BG is the background noise. Fig. 9 The horizontal axis is 1 / T, which represents the adjusted radiation brightness E. i (equivalent to an exposure time of 4096μs).

[0150] [Formula 5]

[0151] E i =(t dst / t d )(I i -I BG ) (9)

[0152] Next, we investigate the blackbody furnace temperature T and the radiation brightness E at each wavelength. i The relationship between the temperature correction coefficient α of formula (10) is obtained by curve regression, as shown in Table 3. i and β i By using this temperature correction coefficient, it is not necessary to use a correction curve for each exposure time, and it is possible to perform a conversion process between the radiance and the temperature for each wavelength using one correction curve.

[0153] [Formula 6]

[0154] E i =α i exp(-β i / T) (10)

[0155] [Table 3]

[0156] symbol Measurement wavelength α β <![CDATA[λ1]]> 1300nm 1.063E+9 1.094E+4 <![CDATA[λ2]]> 1400nm 5.725E+8 1.012E+4 <![CDATA[λ3]]> 1500nm 5.247E+8 9.507E+3 <![CDATA[λ4]]> 1600nm 3.289E+8 8.886E+3

[0157] (Verification of temperature measurement accuracy)

[0158] In order to verify the measurement accuracy of the temperature measurement of the embodiment, the plated original plate was electrically heated to alloy with the Fe substrate, and the thermal radiation light spectrum at this time was measured by using a hyperspectral camera to evaluate the temperature measurement performance of the alloying process. The plated original plate is a steel plate obtained by coating molten zinc on a 0.4 mm thick mild steel plate and solidifying it. Fig.10The structure of the test device is shown in the figure. The plated original plate is heated to the alloying temperature (around 500°C) by electric heating, and alloyed by keeping it at a certain temperature. The radiation intensity during alloying is measured by a hyperspectral camera. At this time, the front and back sides of the measurement point are painted with black body paint, and the radiation thermometer is used to measure the actual temperature, which is compared with the temperature value obtained by the hyperspectral camera.

[0159] (Calculation method of temperature)

[0160] A method for calculating the temperature of a certain position in the plate width direction of the GA steel plate according to this embodiment will be described.

[0161] First, the radiation brightness E of the temperature measurement location is obtained from the hyperspectral image. i Specifically, the wavelength λ at the temperature measurement position is extracted from the obtained two-dimensional hyperspectral image. i The brightness value I of (i=1, 2, 3, 4) i And, through formula (9), the brightness value I i With the reference exposure time t dst =4096μs to obtain the radiant brightness E i .

[0162] Next, calculate the apparent temperature T when the emissivity of the wavelength is equal e and apparent emissivity ε e Specifically, in the calculated radiance E i In the equation, two wavelengths λ are selected. A and B The radiance E A and E B , calculate the temperature and emissivity assuming that the emissivity at the two wavelengths is equal. More specifically, the apparent temperature T e and intensity ratio R(=E A / E B ) is shown in formula (11), so the apparent temperature T can be calculated by formula (12) e Then, using the calculated apparent temperature T e , use formula (13) to calculate the apparent emissivity ε e .

[0163] [Formula 7]

[0164]

[0165] Next, the actual emissivity ε is calculated using the approximate formula (14): A Specifically, let a = A1(1 / λ A -1 / λ B ), b=λB / (λ B -λ A ), use ε e , use formula (14) to calculate ε A Here, A1 uses a value obtained in advance through experiments. In the case of a GA steel plate, A1=0.10.

[0166] [Formula 8]

[0167]

[0168] And, finally, using the calculated radiance ε A Calculate the actual temperature T. Specifically, use the temperature correction coefficient α i and β i , the calculated emissivity ε A and the radiance E A , the temperature T is calculated using equation (16) obtained by transforming equation (15).

[0169] [Formula 9]

[0170] E A =ε A α A exp(-β A / T) (15)

[0171]

[0172] (Calculation results based on approximate formula)

[0173] exist Fig.11 9 shows the results of measuring the brightness values ​​at wavelengths λ1 to λ4 by fixing the temperature at 530° C. and adjusting the exposure time during the alloying process so that the wavelength λ4 with the maximum intensity is not saturated. Fig.11 (a) is the brightness value extracted from the hyperspectral image. Fig.11 (b) is the radiation brightness after background noise correction and conversion to the reference exposure time (4096μs). ET represents the exposure time (unit: μs). I1~14 represents the brightness value of each wavelength. E1~E4 represents the radiation brightness of each wavelength. Fig.11 As shown in (a), the emissivity increases and the brightness value increases during the alloying process, so the exposure time is changed from 4096μs to 2560μs to make Fig.11 The horizontal axis of (a) shows that the radiant brightness does not deviate from the appropriate range at around 41 seconds. As a result, the brightness values ​​at wavelengths λ1 to λ4 decrease in proportion to the change in exposure time. Fig.11In (b), the conversion is performed with the reference exposure time (4096 μs), thereby obtaining a continuously changing brightness value. Therefore, the temperature calculation process based on equation (10) can be performed.

[0174] exist Fig.12 The calculation results of temperature and emissivity are shown in . It can be seen that the emissivity changes from about 0.2 before alloying to about 0.8 after alloying. For reference, Fig.12 In the text, as an embedded thermometer, the temperature measurement results at the same width position (however, on the back side opposite to the surface of the steel plate photographed by the hyperspectral camera) obtained by a previous temperature measurement device (not a device that measures the full width of the steel plate, but a temperature measurement device based on the two-color radiation temperature measurement method with only one position across the width of the steel plate) are recorded. Fig.12 In the case of A1=0.00, which does not consider the wavelength dependence of the emissivity shown in (a), the temperature calculated after alloying is consistent with the actual value measured by the embedded thermometer, but before alloying, the calculated temperature output is about 20°C higher than the temperature measurement result obtained by the embedded thermometer. Fig.12 In the case of the wavelength dependence coefficient A1=0.10 using the GA steel plate shown in (b), the temperature calculated before alloying is consistent with the temperature measurement result obtained by the embedded thermometer, and the temperature calculated during alloying deviates from the temperature measurement result obtained by the embedded thermometer, but the deviation is suppressed to about 5°C. That is, as the wavelength dependence coefficient A1=0.10, it can be seen that the temperature measurement accuracy can be improved by considering the wavelength dependence.

[0175] Thus, it is known that by using the wavelength dependence coefficient to calculate the temperature, the temperature of the coating surface can be measured with high accuracy from before alloying to after alloying. In addition, since it is an approximate calculation, repeated numerical calculations such as convergence calculations are not required, and the temperature can be calculated at high speed.

[0176] (Effect of wavelength switching)

[0177] As described above, it is desirable to adjust the exposure time in accordance with the degree of alloying, but since one imaging element is used to obtain at least a portion of information in the plate width direction of the GA steel plate over a wide range, it is not possible to partially adjust the exposure time for each width direction position in accordance with the alloying unevenness in the plate width direction. Therefore, a method of calculating stable measurement values ​​in both the unalloyed portion and the alloyed portion with the same exposure time was studied.

[0178] exist Fig.13 The figure shows the results of measurements performed at a fixed temperature of 480° C. and at a sensitivity setting (exposure time of about 10000 μs) that ensures sufficient radiance at the temperature before alloying. Fig.13(a) is the brightness value extracted from the hyperspectral image. Fig.13 (b) is the radiant brightness after background noise correction and conversion to the reference exposure time (4096μs). ET represents the exposure time (unit: μs). I1 to I4 represent the brightness values ​​of each wavelength. E1 to E4 represent the radiant brightness of each wavelength. Since the emissivity changes from approximately 0.2 to approximately 0.8 due to alloying, it can be seen that the brightness value at 1600nm on the long wavelength side exceeds the brightness upper limit of 4096 and is saturated. On the other hand, since the brightness value on the short wavelength side has increased due to alloying, a sufficient brightness value for temperature calculation can be obtained.

[0179] exist Fig.14 An example of temperature calculation results is shown in FIG. Fig.14 (a) uses the long wavelength side (1500nm and 1600nm), Fig.14 (b) is the result of calculating the temperature using the radiant brightness on the short wavelength side (1300nm and 1400nm). Fig.14 The temperature measurement results at the same width position (however, the back side opposite to the surface of the steel plate photographed by the hyperspectral camera) obtained by a conventional temperature measurement device (not a device that measures the full width of the steel plate, but a temperature measurement device based on the two-color radiation temperature measurement method at only one position across the width of the steel plate) are recorded as embedded thermometers. If we focus on calculating the temperature deviation (the width from peak to peak of the noise), then Fig.14 As shown in (a) in the figure, the temperature calculation result on the long wavelength side can be accurately calculated before alloying (with a deviation of about 10°C). However, since the imaging element is saturated after alloying, the temperature cannot be calculated. Fig.14 As shown in (b), for the temperature measurement results on the short wavelength side, although they can be stably calculated after alloying (with a deviation of about 10°C), the temperature calculation results have a large deviation (with a deviation of about 20°C) due to the low radiation brightness before alloying. It can be seen that even if the exposure time is the same, as long as the exposure time is controlled so that the radiation brightness of the alloyed part that becomes high on the short wavelength side (1400nm) in the entire width direction of the plate is not saturated, the temperature can be calculated stably and accurately in the width direction of the GA steel plate by using the wavelength set on the short wavelength side for the alloyed part with higher emissivity and using the wavelength set on the long wavelength side for the pre-alloyed part with lower emissivity.

[0180] (How to adjust the exposure time)

[0181] As described above, even at any position in the width direction of the board within the measurement object range, it is necessary to meet the condition that at least one set of wavelengths can detect the radiant brightness from the minimum brightness to the upper limit of the brightness. In order to ensure effective measurement with a certain wavelength combination (λ1 and λ2, λ2 and λ3, λ3 and λ4) in the entire width direction of the board, the exposure time of the hyperspectral camera is controlled.

[0182] First, a hyperspectral image is obtained, and the left and right edge positions of the GA steel plate are detected based on the hyperspectral image, so as to grasp the range where the GA steel plate exists on the hyperspectral image. As a method for detecting the edge position, a method such as differentiating the distribution of the radiant brightness in the plate width direction and exploring its maximum and minimum values ​​can be used.

[0183] Next, regarding wavelength λ3, we search for the minimum radiance I within the measurement target range. 3min , calculate the brightness not less than the lower limit I' min The critical maximum exposure time D max =I' min / I 3min Dp. Dp is the exposure time of the acquired hyperspectral image. Next, regarding wavelength λ2, the maximum radiance I within the measurement target range is searched. 2max , calculate the brightness limit I' max The critical minimum exposure time D min =I' max / I 2max ·Dp. Next, select min To D max Change the exposure time of the hyperspectral camera. min To D max If there is no exposure time that can be set for the hyperspectral camera, the exposure time closest to the exposure time that can be set is selected.

[0184] (Measurement result example)

[0185] For the obtained hyperspectral image, the brightness values ​​I1~I4 at each position in the width direction of the plate are observed, and a wavelength combination (λ1 and λ2, λ2 and λ3, λ3 and λ4) that obtains effective brightness is selected, and the temperature is measured using the two-color radiation temperature measurement method.

[0186] In order to accurately calculate the temperature and emissivity using the two-color radiation temperature measurement method, the detection brightness under the two wavelengths used needs to be large relative to the measurement resolution and not saturated. That is, the brightness lower limit I' is preset min And the upper limit of brightness I' max In order to achieve an appropriate brightness range, when using a 12-bit hyperspectral camera, the lower limit of brightness is I'min After background noise correction, it is about 500, and the upper limit of brightness is I' max It is about 3500. Since the hyperspectral camera can detect the radiation brightness at multiple wavelengths, when calculating the temperature at two wavelengths, the combination of wavelengths is selected based on the detected brightness at the two wavelengths.

[0187] By using a hyperspectral camera and speeding up the computational processing using approximate equations, it is possible to output 20 width-direction temperature measurement results per second. Fig.15 shows a measurement example in which alloying is uneven in the width direction. By measuring the temperature and emissivity across the width direction of the GA steel plate, it is possible to clearly confirm that alloying is delayed near the edge of the GA steel plate due to low temperature. In this way, the temperature distribution meter can grasp the alloying state in addition to the temperature distribution in the width direction, so it is effective for improving the manufacturing conditions during the alloying process of the GA steel plate.

[0188] In addition, in the above description, the case where the temperature of the GA steel plate is mainly measured is mainly described as an example, but the technology disclosed in the present invention is not limited to the GA steel plate, and can be applied to all steel plates. In particular, when the spectral emissivity of the steel plate varies greatly depending on the situation, the technology disclosed in the present invention can be appropriately applied.

[0189] An example of the technology of the present disclosure has been described above, but the technology of the present disclosure is not limited to the above and can of course be implemented with various modifications other than the above without departing from the gist of the invention.

[0190] The entire contents disclosed in Japanese Patent Application No. 2022-125745 are incorporated herein by reference.

[0191] Description of symbols

[0192] 10, 210 temperature measuring device; 12 camera unit; 14, 214 operation processing unit; 22, 222 selection unit; 24, 222 calculation unit; 26 setting unit; 32 CPU; 34 memory; 36 storage device; 38 input device; 40 output device; 42 storage medium reading device; 44 communication I / F; 46 bus.

Claims

1. A temperature measuring device for measuring the temperature of a steel plate. The temperature measuring device has: As an imaging unit of a spectroscopic camera, a range including at least a portion of the steel plate in the plate width direction is taken as a field of view, and the radiant brightness of the steel plate under multiple wavelengths is obtained at each position on the surface of the steel plate within the field of view; and The calculation processing unit selects two wavelengths from the wavelengths of the radiation brightness of the above-mentioned steel plate obtained by the above-mentioned camera unit, and uses the radiation brightness of the above-mentioned steel plate obtained by the above-mentioned camera unit at the two selected wavelengths to calculate the temperature of each position on the surface of the above-mentioned steel plate through two-color radiation temperature measurement.

2. The temperature measuring device according to claim 1, The arithmetic processing unit holds at least two wavelength sets, the wavelength set consisting of two wavelengths among the wavelengths of the radiant brightness of the steel plate obtained by the imaging unit, When the position of the surface of the steel plate whose temperature is to be measured is in an area where the radiation brightness of the steel plate is large, the calculation processing unit selects two wavelengths included in a set of two wavelengths on the short wavelength side in the plurality of wavelength sets maintained; when the position of the surface of the steel plate whose temperature is to be measured is in an area where the radiation brightness of the steel plate is small, the calculation processing unit selects two wavelengths included in a set of two wavelengths on the long wavelength side in the plurality of wavelength sets maintained. The arithmetic processing unit calculates the temperature of each position on the surface of the steel plate by two-color radiation temperature measurement using the radiation brightness of the steel plate acquired by the imaging unit at the two selected wavelengths.

3. The temperature measuring device according to claim 2, The temperature measuring device is a device for measuring the temperature of the alloyed hot-dip galvanized steel sheet as the steel sheet. When the position of the surface of the steel plate whose temperature is to be measured is in a region where alloying of the surface of the steel plate progresses smoothly and the radiation brightness of the steel plate is large, the arithmetic processing unit selects two wavelengths included in a set of two wavelengths on the short wavelength side combined from among the plurality of wavelength sets maintained, When the position of the surface of the steel plate whose temperature is to be measured is in an area where alloying of the surface of the steel plate has not progressed and the radiation brightness of the steel plate is small, the calculation processing unit selects two wavelengths included in a set of two wavelengths on the long wavelength side from the multiple wavelength sets maintained.

4. The temperature measuring device according to claim 2 or 3, When the radiant brightness of the steel plate obtained by the wavelength set including the longest wavelength among the plurality of wavelength sets held by the calculation processing unit is lower than a preset lower limit threshold, the calculation processing unit increases the exposure amount of the imaging unit, When the radiant brightness of the steel plate obtained with the wavelength set including the shortest wavelength among the plurality of wavelength sets stored in the calculation processing unit is higher than a preset upper limit threshold, the calculation processing unit reduces the exposure amount of the imaging unit.

5. The temperature measuring device according to any one of claims 1 to 4, The arithmetic processing unit divides the field of view of the imaging unit into a plurality of partitions along the width direction of the steel plate, and obtains a wavelength range of the radiation brightness of the steel plate successfully obtained by the imaging unit for each of the partitions. The arithmetic processing unit selects two wavelengths from a range of wavelengths in which the radiant brightness of the steel plate is successfully obtained for each of the partitions. The arithmetic processing unit calculates the temperature of each position on the surface of the steel plate by two-color radiation temperature measurement using the radiation brightness of the steel plate acquired by the imaging unit at the two selected wavelengths.

6. The temperature measuring device according to any one of claims 1 to 5, The arithmetic processing unit selects wavelengths shorter than a wavelength at which the radiance of the steel sheet is maximum as the two wavelengths.

7. The temperature measuring device according to any one of claims 1 to 6, The arithmetic processing unit detects edge positions of both sides of the steel plate in a plate width direction based on the image acquired by the imaging unit, and determines a range for measuring the temperature based on the detected edge positions.

8. A temperature measurement method for measuring the temperature of a steel plate. This temperature measurement method has: a brightness acquisition step, using an imaging unit as a spectroscopic camera to acquire the radiant brightness of the steel plate at multiple wavelengths at various positions on the surface of the steel plate within a field of view, wherein the imaging unit defines a range including at least a portion of the steel plate in a plate width direction as the field of view; and The temperature calculation step uses a calculation processing unit to select two wavelengths from the wavelengths of the radiation brightness of the steel plate obtained by the above-mentioned camera unit, and uses the radiation brightness of the above-mentioned steel plate obtained by the above-mentioned camera unit at the two selected wavelengths to calculate the temperature of each position on the surface of the above-mentioned steel plate through two-color radiation temperature measurement.

9. The temperature measurement method according to claim 8, The arithmetic processing unit holds at least two wavelength sets, the wavelength set consisting of two wavelengths among the wavelengths of the radiant brightness of the steel plate obtained by the imaging unit, In the temperature calculation step, when the position of the surface of the steel plate whose temperature is to be measured is in an area where the radiation brightness of the steel plate is large, two wavelengths included in a set of two wavelengths on the short wavelength side in the plurality of wavelength sets retained are selected; when the position of the surface of the steel plate whose temperature is to be measured is in an area where the radiation brightness of the steel plate is small, two wavelengths included in a set of two wavelengths on the long wavelength side in the plurality of wavelength sets retained are selected. In the temperature calculation step, the temperature of each position on the surface of the steel plate is calculated by two-color radiation temperature measurement using the radiation brightness of the steel plate acquired by the imaging unit at the two selected wavelengths.

10. The temperature measurement method according to claim 9, The temperature measurement method is a method of measuring the temperature of the alloyed hot-dip galvanized steel sheet as the steel sheet. In the temperature calculation step, when the position of the surface of the steel plate whose temperature is to be measured is in a region where alloying of the surface of the steel plate progresses smoothly and the radiation brightness of the steel plate is large, two wavelengths included in a set of two wavelengths on the short wavelength side combined from among the plurality of wavelength sets retained are selected, In the temperature calculation step, when the position of the surface of the steel plate whose temperature is to be measured is in an area where alloying of the surface of the steel plate has not progressed and the radiation brightness of the steel plate is small, two wavelengths included in a set combining two wavelengths on the long wavelength side are selected from the plurality of wavelength sets retained.

11. The temperature measurement method according to claim 9 or 10, In the temperature calculation step, when the radiation brightness of the steel plate obtained by the wavelength set including the longest wavelength among the plurality of wavelength sets held by the calculation processing unit is lower than a preset lower limit threshold, the exposure amount of the imaging unit is increased, In the temperature calculation step, when the radiation brightness of the steel plate obtained by the wavelength set including the shortest wavelength among the plurality of wavelength sets held by the calculation processing unit is higher than a preset upper limit threshold, the exposure amount of the imaging unit is reduced.

12. The temperature measurement method according to any one of claims 8 to 11, In the temperature calculation step, the field of view of the imaging unit is divided into a plurality of partitions along the width direction of the steel plate, and a wavelength range of the radiation brightness of the steel plate successfully obtained by the imaging unit is obtained for each of the partitions. In the temperature calculation step, two wavelengths are selected from a range of wavelengths at which the radiant brightness of the steel plate is successfully obtained for each of the partitions. In the temperature calculation step, the temperature of each position on the surface of the steel plate is calculated by two-color radiation temperature measurement using the radiation brightness of the steel plate acquired by the imaging unit at the two selected wavelengths.

13. The temperature measurement method according to any one of claims 8 to 12, In the temperature calculation step, wavelengths shorter than the wavelength at which the radiance of the steel plate is maximum are selected as the two wavelengths.

14. The temperature measurement method according to any one of claims 8 to 13, In the temperature calculation step, edge positions of both sides of the steel plate in the plate width direction are detected based on the image acquired by the imaging unit, and a range for measuring the temperature is determined based on the detected edge positions.

Images