Temperature measuring device, temperature measuring method, and program

Abstract

To provide a temperature measuring device, a temperature measuring method, and a program with which it is possible to measure a two-dimensional temperature distribution of a measurement object at low cost, raise sensor output for low-temperature objects whose amount of radiation is small, and raise temperature measurement accuracy for high-temperature objects whose amount of radiation is large.SOLUTION: The temperature measuring device comprises: a single image sensor of global shutter type including pixels in two dimensions; a rotary filter in which a first filter region and a second filter region that first light and second light differing in wavelength pass through, respectively, are provided in plurality in order; light guide means for guiding light from a measurement object to the image sensor; and an output addition unit for adding the output of first light having passed through the plurality of first filter regions in pixel units and adding the output of second light having passed through the plurality of second filter regions in pixel units. The temperature of the measurement object is measured on the basis of the output of the output addition unit or the arithmetic mean of the outputs of the output addition unit.SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] The present invention relates to a temperature measurement device, a temperature measurement method, and a program that can measure the two-dimensional temperature distribution of an object at low cost without requiring a complex optical system or image sensor, and that can increase sensor output for low-temperature objects with low radiation levels and improve temperature measurement accuracy for high-temperature objects with high radiation levels. [Background technology]

[0002] Known temperature measurement devices for measuring the temperature of a measurement object use the two-color temperature method, which converts the temperature from the ratio of radiance using two different wavelengths that are close to each other. Among various specific methods using the two-color temperature method, one known from Patent Document 1 measures temperature at low cost by rotating filters with different transmission wavelength ranges that are installed in the optical path from the measurement object to the sensor. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Special Publication No. 61-011368 Summary of the Invention [Problem to be solved by the invention]

[0004] However, Patent Document 1 not only fails to measure the two-dimensional temperature distribution of the object, but also does not take into consideration the convenience of increasing sensor output for low-temperature objects with low radiation levels, or increasing temperature measurement accuracy for high-temperature objects with high radiation levels.

[0005] The object of the present invention is to provide a temperature measurement device, a temperature measurement method, and a program that can measure the two-dimensional temperature distribution of an object at low cost, increase sensor output for low-temperature objects with low radiation, and increase temperature measurement accuracy for high-temperature objects with high radiation. [Means for solving the problem]

[0006] In order to achieve the above object, the temperature measuring device of the present invention comprises a single image sensor having two-dimensional pixels and employing a global shutter system in which each pixel is simultaneously exposed to light and outputs a digital signal; a rotary filter having multiple sets of first and second filter regions arranged in order, through which first light and second light of different wavelengths pass, respectively; a light guiding means for guiding the first light and the second light from an object to be measured to the image sensor in a time-series manner via the rotary filter; and an output adder for adding together the outputs of the first light that have passed through multiple first filter regions in units of each pixel, and adding together the outputs of the second light that have passed through multiple second filter regions in units of each pixel, and is characterized in that the temperature measuring device measures the temperature of the object to be measured based on the output of the output adder or an averaging of the outputs of the output adder.

[0007] Furthermore, a temperature measurement method according to the present invention is characterized by comprising the steps of: providing a single image sensor in a light path from an object to be measured, the image sensor having two-dimensional pixels and a global shutter system in which each pixel is simultaneously exposed to light and outputs a digital signal; providing a rotary filter corresponding to the image sensor, the rotary filter having a plurality of sequentially arranged first and second filter regions through which first light and second light having different wavelengths pass; guiding the first light and the second light from the object to be measured to the image sensor via the rotary filter in a time series manner; adding together the outputs of the first light that have passed through the plurality of first filter regions in units of each pixel and adding together the outputs of the second light that have passed through the plurality of second filter regions in units of each pixel; and measuring the temperature of the object to be measured based on the output of the output adder or the averaging of the outputs of the output adder.

[0008] The present invention also provides a program for causing a computer to function as each of the means of the temperature measuring device. [Effects of the Invention]

[0009] According to the present invention, the two-dimensional temperature distribution of the object can be measured at low cost, and the sensor output can be increased for low-temperature objects with low radiation, and the temperature measurement accuracy can be improved for high-temperature objects with high radiation. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a diagram illustrating an overall configuration of an embodiment of the present invention. [Figure 2] FIG. 1 is a block diagram of an embodiment of the present invention. [Figure 3] 1A and 1B are diagrams showing the configuration of a rotary plate equipped with different filters according to an embodiment of the present invention, in which (a) shows a two-wavelength, four-division configuration, and (b) shows a two-wavelength, 20-division configuration. [Figure 4] FIG. 10 is an explanatory diagram of a synchronization signal generating system according to an embodiment of the present invention in which a rotating plate does not have a mark for forming a synchronization signal. [Figure 5] FIG. 10 is a diagram illustrating an output of a trigger sensor according to an embodiment of the present invention. [Figure 6] 1 is a flowchart according to an embodiment of the present invention. [Figure 7] 10A to 10C are explanatory diagrams of image addition processing and averaging processing of image addition processing according to an embodiment of the present invention. [Figure 8] FIG. 10 is an explanatory diagram of a second embodiment in which a rotating plate is provided with marks for forming synchronization signals. [Figure 9] FIG. 10 is an explanatory diagram of a modified example in which a measuring head as a measurement terminal is connected to a server via the Internet. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

[0012] (First embodiment) Prior to a specific description of this embodiment, the two-color thermometry method will be described first.

[0013] (Two-color thermometry) The concept of two-color thermometry is explained below. To measure the temperature of high-temperature workpieces above 300°C without contact, the electromagnetic radiation emitted by the object is measured and the temperature is calculated from its intensity. This method is commercially available as a radiation thermometer, which obtains temperature from visible or infrared light, or as a thermograph, which measures temperature distribution. The object is generally a non-blackbody, and to obtain the true temperature from a radiation thermometer or a thermograph, correction must be made for the emissivity, which is the ratio of the amount of radiation to that of a blackbody. However, the emissivity of a non-blackbody generally varies depending on its material, shape, and temperature, making it difficult to obtain an accurate temperature.

[0014] Therefore, two-color thermometry is used. Two-color thermometry focuses on the fact that the emissivity of radiation at two closely spaced wavelengths is the same, and since the ratio of the radiation amounts at the two wavelengths is functionally related to the true temperature, the radiation amounts at the two wavelengths are measured to calculate the true temperature. This can be explained using the following formula:

[0015] The Wien radiation formula for the amount of radiation Mλ at wavelength λ is:

[0016]

number

[0017] where c1: First radiation constant = 2πc2h = 3.741844 x 10-16 [Wm2] λ: Wavelength μm c2: Second radiation constant = 1.438769X10-2[m K] T: Absolute temperature (Kelvin) Let the two wavelengths to be measured, λ1 and λ2, their emissivities, ε1 and ε2, and the radiation amounts of both, M1 and M2, respectively. Let R be the ratio of the radiation amounts at both wavelengths (output ratio at the image sensor). R can be expressed by equation (2).

[0018]

number

[0019] When the condition ε1=ε2 is met, equation (2) becomes equation (3).

[0020]

number

[0021] By rearranging both sides, the temperature T can be obtained by equation (4).

[0022]

number

[0023] Here, C3 and C4 are constants calculated from both wavelengths and are expressed by the following formula.

[0024]

number

[0025] Even if the emissivity at two wavelengths or the transmittance of an inclusion differs, if the ratio does not change with temperature, the true temperature of an object whose temperature changes can be determined by correcting the ratio with a black body.

[0026] (Rotating filter type temperature measurement) Figure 1 shows the overall configuration of a temperature measurement device according to an embodiment of the present invention, which is composed of a measurement head and a computer. The measurement head includes an objective lens 1, a rotary filter 2, an image sensor unit 3, a wavelength-specific image distribution circuit 4, an image addition / non-addition processing circuit 5, a memory bank 6, a trigger light source 8, a trigger sensor 9, and a motor 10 for rotating the rotary filter 2 at a constant speed. A computer 7, which specifies exposure time and processes two-color temperatures, is equipped with computer hardware and software. The image sensor unit 3 includes an image sensor 3a, an analog amplifier 3b, and an A / D conversion element 3c.

[0027] These components are numbered the same as in the block diagram of FIG.

[0028] In this embodiment, in order to reduce costs, a rotary filter 2 is provided between the objective lens 1 and the image sensor 3a. The rotary filter 2 functions as a bandpass filter, and is provided with a first filter region and a second filter region adjacent to each other in the rotational direction, through which a first light and a second light having different wavelengths pass, respectively, and multiple sets of these filter regions are provided in the rotational direction.

[0029] In Fig. 3(a), the filter is divided into four parts, with two sets of a first filter 12 and a second filter 13. In Fig. 3(b), the filter is divided into 20 parts, with ten sets of a first filter 22 and a second filter 23.

[0030] While conventional two-color thermometers measure one point or one area in one dimension, this embodiment uses an image sensor and measures a large number of points, such as 300,000 points (640x480).

[0031] Furthermore, while conventional two-color thermometers rely on analog signal processing with floating elements to measure temperature from the radiation ratio of two wavelengths, this embodiment is able to perform accurate numerical processing based on Wien's radiation law to calculate the temperature of each of the approximately 300,000 pixels.

[0032] In this embodiment, the image sensor 3a is a single image sensor with two-dimensional pixels and a global shutter in which each pixel is simultaneously exposed to light and outputs a digital signal. The global shutter method is adopted because, when performing dynamic imaging with the rotary filter 2 interposed, it is preferable that each pixel be simultaneously exposed to light and output a digital signal.

[0033] The objective lens 1 captures an image of the object to be measured on the image sensor 3a through a bandpass filter configured on the disk of the rotary filter 2. In this way, the objective lens 1 functions as a light guiding means that guides the first light and second light from the object to the image sensor 3a in time series via the rotary filter 2. As a light guiding means, a glass fiber image guide tube is provided on the focal plane of the objective lens, and a lens is placed at the end of the image guide fiber, which transmits the light to the image sensor through a filter disk. This makes it possible to perform temperature measurement while avoiding stray light from around the object to be measured.

[0034] Here, prior to temperature measurement, the computer 7 instructs the image sensor unit 3 to set an exposure time, and the image sensor 3a can set the exposure time in the global shutter function (for example, if step S1 in Figure 6 described later is YES).

[0035] In the image sensor unit 3, the output of each pixel of the image sensor 3a is amplified by the analog amplifier 3b, and then sent as 12- to 16-bit digital data by the A / D converter 3c to the wavelength-specific image distribution circuit 4. The images are then stored in each wavelength region of the memory bank 6 as images of wavelengths identified by the trigger signal, which is the output of the trigger light-receiving sensor 9 (Fig. 2).

[0036] Regarding data reading, the image adding / non-adding processing circuit 5 reads the required two pairs of wavelength data from the memory bank 6, and after adding or non-adding the data for each pixel according to a command from the computer 7, outputs it to the computer 7. The computer 7 then performs two-color temperature processing on the non-added or added image data, and then outputs a signal for display or control.

[0037] Regarding image addition in the image addition / non-addition processing circuit 5, when the rotation speed of the disk of the rotary filter 2 is P revolutions per second (RPS) and the number of filter divisions is Q, if the required exposure time is longer than 1 / (P × Q) seconds, two pairs of memory values ​​read from the memory bank 6 are added together (Figure 7). For example, if the rotation speed of the disk of the rotary filter 2 is 2400 RPM (40 revolutions per second) and the number of filter divisions is 4 (the disk is divided into four), the exposure time per wavelength is 1 / 40 × 1 / 4 × 1000 = 6.25 milliseconds (ms), but if two images for each wavelength are added together, the exposure time per wavelength can be extended to 12.5 milliseconds (ms).

[0038] (Sync signal detection) In a rotating filter, a synchronization signal with the filter position is required so that the image sensor can capture an image when the filter area that transmits a specific wavelength is in front of the image sensor.

[0039] In this embodiment, as shown in FIGS. 1 and 4, light from a trigger light source 8 is received by a trigger light receiving sensor (photodiode) 9 via a trigger filter 50 (FIG. 4), and the output thereof is used.

[0040] That is, since the trigger filter 50 is the same filter (same spectral transmittance characteristics) as one of the different filters of the rotary filter 2, the output of the trigger light-receiving sensor 9 is turned on when the filter on the optical path of the rotary filter 2 is the same as the trigger filter 50. On the other hand, when the filter on the optical path of the rotary filter 2 is different from the trigger filter 50, the output of the trigger light-receiving sensor 9 is turned off.

[0041] In this embodiment, the synchronization signal can be detected in this manner, and the image signals can be sorted by wavelength by the wavelength-specific image distribution circuit 4 and stored in a predetermined memory area of ​​the memory bank 6 (FIG. 2).

[0042] In this embodiment, regardless of the number of divisions (even if the number of filter divisions is increased), there is no need to give up the trigger detection area to the rotary filter 2, and a wide filter area can be used.

[0043] The output of the trigger sensor in this case is shown in Figure 5. Image sensor 14a transitions from an output state where light of the first wavelength λ1 is received (ON) and light of the second wavelength λ2 is not received (OFF) to an output state where light of the first wavelength λ1 is not received (OFF) and light of the second wavelength λ2 is received (ON).

[0044] In this embodiment, the above-mentioned synchronization signal (trigger signal) is obtained, and the image signals are sorted by wavelength using the signal and stored in a predetermined memory area (memory bank 6) shown in FIG. Light of wavelength λ1 that passes through the filter is stored in memory areas 1-1, 1-2, ... 1-n in sequence as the rotary filter 2 rotates. Also, light of wavelength λ2 that passes through the λ2 filter is stored in memory areas 2-1, 2-2, ... 2-n in sequence as the rotary filter 2 rotates.

[0045] (Time delay between two wavelength images and input time to the image sensor depending on the number of filter divisions and rotation speed) In this embodiment, the time (exposure time) for receiving light emitted from the measurement target in the image sensor can be precisely set. For example, if a disk is divided into four parts and two wavelengths are assigned to each of the two wavelength filters, and the disk is rotated at 1500 RPM (25 rotations per second), the time required to input one image to the image sensor (exposure time) and the time delay between the two wavelength images are each 1 / 25 × 1 / 4 × 1000 = 10 milliseconds (ms).

[0046] 1) When the temperature is high and the radiation is strong (step S1 or S2 in Figure 6 is YES) If the temperature of the object to be measured is high and the radiation intensity is strong, changing the motor rotation speed from the initial standard value of 750 RPM (12.5 rotations per second) to 3000 RPM (50 rotations per second) will reduce the exposure time and the time delay between the two wavelength images (λ1 to λ2 time difference) to 1 / 50 x 1 / 4 x 1000 = 5 milliseconds. The exposure time will be 1 / 2 of the initial default value of 10 milliseconds.

[0047] The exposure time can be shortened by increasing the rotations per minute even further, but this embodiment, which uses a sensor with a global shutter function as the image sensor, can shorten the exposure time to about 1 microsecond. However, in this case, the time delay between the two wavelength images (λ1 to λ2 time difference) cannot be eliminated (see Table 1 below). Increasing the motor's rotations per minute to shorten this radiation capture time can cause noise and vibration, and durability must also be considered.

[0048] Therefore, as a way to shorten the time delay between the two wavelength images without increasing the rotation speed of the motor, it is possible to deal with this by increasing the number of divisions in the rotating disk filter. For example, if the number of filter divisions is set to 20, i.e., 20 divisions (Figure 3(b)), the exposure time at 3000 RPM and the time delay between the two wavelength images will be 1 / 50 × 1 / 20 × 1000 = 1 millisecond, respectively (see Table 2 below).

[0049] Here, the time delay between the two wavelength images cannot be determined solely by the rotation speed of the rotating filter, but is limited by the FPS (number of images taken per second, i.e., shooting speed) of the image sensor. At 30 FPS, which is the normal shooting speed, the time delay between the two wavelength images is 30.3 ms, but to reduce the time delay to 1 ms, high-speed shooting of 1000 FPS is required.

[0050] 2) When the temperature is low and the radiation amount is weak (steps S1 and S2 in Figure 6 are NO) When the temperature of the object being measured is low and the radiation intensity is weak, which corresponds to the case where a long exposure time is required, the motor rotation speed should be slowed. For example, if the 4-segment filter disk is rotated at 750 RPM (12.5 revolutions per second), the exposure time for one image and the time delay between the two wavelength images will be 1 / 12.5 x 1 / 4 x 1000 = 20 milliseconds (ms). If the motor rotation speed is further slowed to 150 RPM (2.5 revolutions per second), the time delay becomes 1 / 2.5 x 1 / 4 x 1000 = 100 milliseconds (ms). This is fine for objects that do not change due to flow, but is not suitable for objects that do change due to flow.

[0051] As a solution to this problem, the motor rotation speed remains constant (the standard value of 1500 RPM), and the data at the same address for each pixel of the images captured in Memory Bank 6 (Figure 2) for each wavelength is added together (image addition) to increase the exposure time. For example, if a rotating filter divided into four sections and assigned two wavelengths to each of the two wavelength filters is rotated at 1500 RPM (25 rotations per second), the time delay between the input of one image to the image sensor and the two-wavelength images is 1 / 25 × 1 / 4 × 1000 = 10 milliseconds (ms). Adding 10 images of each wavelength results in an exposure time of 100 ms (see Table 3 below). In this case, the time delay between the two wavelengths is the same 100 ms as when the motor rotation speed is set to 150 RPM (2.5 rotations per second), but the advantage is that the motor rotation speed is 10 times faster (1500 RPM), allowing the exposure time to be changed with minimal rotational fluctuations.

[0052] Furthermore, if a rotating filter divided into 20 parts, each with 10 wavelengths assigned to each of the two wavelength filters, is rotated at 750 RPM (12.5 revolutions per second), the time delay between the input of one image to the image sensor and the two wavelength images is 1 / 12.5 x 1 / 20 x 1000 = 4 milliseconds, and adding together 16 images of each wavelength results in an exposure time of 64 milliseconds (see Table 4 below).

[0053] Furthermore, with regard to image addition, if you want to reduce noise components such as dust particles floating between the measurement target and the objective lens or dynamic noise from the sensor, you can address this by increasing the number of divisions in the rotating filter and adding the data from each pixel (image addition).In this case, for example, if you replace the 4-division filter disk with an 8-division filter disk and rotate it at 1500 RPM (25 rotations per second), the time delay between the input time of one image to the image sensor and the two-wavelength images is 1 / 25 × 1 / 8 × 1000 = 5 milliseconds (ms), and adding 10 images of each wavelength results in an exposure time of 50 ms.

[0054] [Table 1]

[0055] Filter values ​​when divided into 4 sections (rotation speed changed) RPM = revolutions per minute, RPS = revolutions per second, FPS = images per second

[0056] [Table 2]

[0057] Values ​​when the filter is divided into 20 parts (rotation speed changed) RPM = revolutions per minute, RPS = revolutions per second, FPS = images per second

[0058] [Table 3]

[0059] Filter 4 division values ​​(addition number changed at 1500 RPM)

[0060] [Table 4]

[0061] Values ​​when the filter is divided into 20 parts (addition number changed at 750 RPM)

[0062] (Flowchart for temperature measurement) Next, a flowchart relating to temperature measurement will be described with reference to FIG.

[0063] 1) Initial setup and flow overview As the initial settings, the motor rotation speed is set to 750 RPM and the image sensor exposure time is set to 10 milliseconds. In this state, the measurement head is pointed at the measurement target (initial settings).

[0064] If the image displayed on the monitor at the initial setting is deemed unsatisfactory but visible (if step S1 in Figure 6 is YES), the objective lens is adjusted to bring it into focus. The image sensor exposure time is then set to the initial setting of 10 milliseconds. In step S4, the image averaging process shown in Figure 7 (described later) is used to improve the accuracy of temperature measurement.

[0065] If the image is overexposed and blown out in the initial settings (if step S2 in Figure 6 is YES), the exposure time of the image sensor is adjusted. When a measurement command is issued in this state, the image signal from the sensor unit is stored in the λ1 or λ2 image memory without undergoing addition processing. The two-color temperature processing unit reads the image data of both temporally adjacent wavelengths, performs two-color temperature processing, and then displays the temperature or sends a control signal to an external device.

[0066] If the sensor shutter setting exceeds the limit of exposure time reduction control, set the motor rotation speed (RPM) higher and then set the sensor exposure time.

[0067] Then, in step S4, an arithmetic average of image addition as shown in FIG. 7, which will be described later, is used to improve the accuracy of temperature measurement.

[0068] If the image displayed on the monitor by default is dark due to insufficient exposure (if step S2 in Figure 6 is NO), multiple images of the same wavelength are added pixel by pixel in chronological order (image addition) to create an image density that allows two-color temperature measurement. In this case, dynamic noise is reduced by adding temporally consecutive images rather than adding a single image. The number of images required to be added is calculated from the initial image, and the computer issues an instruction to the image addition / non-addition processing circuit to add the images, which are then stored in image memory. The two-color temperature processing unit reads image data of both temporally adjacent wavelengths, processes them into two-color temperature data, and then displays the temperature or sends a control signal to an external device.

[0069] 2) Specific flow First, in step S1, it is determined whether the measurement object is visible (thermal radiation from the measurement object is light in the visible region).

[0070] 1) If the result of step S1 is YES, a commercially available photodiode array, CCD, CMOS, or other image sensor is used to measure high temperatures above 700°C, and the exposure time for a single image sensor using a global shutter is set. Then, the process moves to step S4, where the image is added and averaged (to improve the accuracy of temperature measurement). For convenience, step S4 will be explained in 2-2) below.

[0071] After the image averaging process (to improve the temperature measurement accuracy) is performed in step S4, the output of the image sensor is stored as an image in a memory address determined in response to the trigger signal in step S5.

[0072] 2) In step S1, if the answer is NO (the target is not visible by default), for temperature measurement below 700°C, the amount of thermal radiation is small and the central wavelength range of the radiation is in the near-infrared range, which is longer than visible light. Therefore, it is not possible to receive the light with a silicon-based C-MOS color image sensor or the like, and an image sensor made of a material such as InGaAs that is sensitive to the near-infrared range must be used.

[0073] If NO in step S1, it is determined in step S2 whether the RPM should be increased (whether the rotary filter 2 can be rotated at high speed). If YES in step S2, the temperature is measured at a medium to high temperature of 500°C or higher but less than 700°C, and if NO in step S2, the temperature is measured at a low temperature of less than 500°C.

[0074] 2-1) If the result in step S2 is NO, the RPM is set to slow down the rotating filter 2 as a low-temperature measurement of less than 500°C and more than 300°C, and the process moves to step S3, where the number of added images is calculated. Regarding the image addition process, as shown in Figure 7, the same pixel outputs (outputs of the same coordinate values) S1, S2, ... S of the sensor that receives light that has passed through the λ1 filter are added. n Similarly, the same pixel outputs (outputs of the same coordinate values) T1, T2, ...T of the sensor that receives light that has passed through the λ2 filter are added (image addition). n The temperature is measured based on the ratio of the output of both images.

[0075] Generally, when measuring the temperature of a low-temperature object with little radiation, the sensor output can be increased by using this type of image addition processing.

[0076] 2-2) If the result of step S2 is YES, the RPM of the rotary filter 2 is set to a high speed for measuring medium to high temperatures between 500°C and 700°C, and a short exposure time is set as the sensor exposure time. Then, in step S4, the same pixel outputs (outputs of the same coordinate values) S1, S2, ... S of the sensor that receives light transmitted through the λ1 filter as shown in FIG. n Similarly, the same pixel output (output of the same coordinate value) T1, T2, ... T of the sensor that receives light that has passed through the λ2 filter is calculated. n The average is calculated by adding the values ​​and dividing by n. The temperature is then measured based on the ratio of the two averages.

[0077] Generally, when measuring the temperature of high-temperature (medium-high temperature) objects that emit a lot of radiation, using this type of image addition averaging process can compensate for the decrease in temperature measurement accuracy caused by floating particles such as dust in the optical path and dynamic noise from the sensor and analog electronic components (improving temperature measurement accuracy).

[0078] In the above-described embodiment, as a temperature measurement device used when the object to be measured is a low-temperature object not exceeding 500°C, the temperature of the object to be measured can be measured based on the output of the output addition unit used to increase the output of the image sensor.

[0079] In addition, as a temperature measurement device used when the object to be measured is a high-temperature object exceeding 500°C, the temperature of the object to be measured can be measured based on the arithmetic average of the outputs of the output addition unit used to improve the accuracy of temperature measurement.

[0080] In addition, as a temperature measurement device that can be used when the object to be measured is a low-temperature object that does not exceed 500°C, as well as when the object to be measured is a high-temperature object that exceeds 500°C, it can measure the temperature of the object to be measured based on the output of the output adder that is used to increase the output of the image sensor when the object to be measured is a low-temperature object that does not exceed 500°C, and can measure the temperature of the object to be measured based on the arithmetic average of the outputs of the output adder that is used to increase the accuracy of temperature measurement when the object to be measured is a high-temperature object that exceeds 500°C.

[0081] In addition, when the object to be measured is a low-temperature object that does not exceed 500°C, the control unit can make the rotation speed of the rotary filter slower than a reference value (e.g., 750 RPM or 2000 RPM), and when the object to be measured is a high-temperature object that exceeds 500°C, the control unit can make the rotation speed of the rotary filter faster than the above-mentioned reference value.

[0082] Furthermore, when the measurement target is a high-temperature object exceeding 500° C., the control unit can make the exposure time of the image sensor shorter than a predetermined value (for example, 10 milliseconds).

[0083] (Second embodiment) A rotary filter according to a second embodiment of the present invention is shown in FIG. 8 together with an image sensor 14b. The rotary filter of this embodiment is divided into four sections, with two sets of a first filter 32 and a second filter 33. This rotary filter is provided with synchronization detection marks 16 to 19. A light source and a sensor for reading these synchronization detection marks are provided on either side of the disk of the rotary filter. The trigger light source 8 and trigger detection sensor 9 described in the first embodiment in FIG. 4 can be used, with the trigger filter 50 removed.

[0084] This embodiment is the same as the first embodiment except for the rotary filter and the synchronous detection system.

[0085] (Temperature measurement method) In addition to the temperature measuring device according to the above-described embodiment, the present invention can also be expressed as a temperature measuring method, comprising the steps of: providing a single image sensor in a light path from a measurement object, the image sensor having two-dimensional pixels and a global shutter system in which each pixel is simultaneously exposed to light and outputs a digital signal; providing a rotary filter corresponding to the image sensor, the rotary filter including a plurality of sequentially arranged first and second filter regions through which first and second light beams of different wavelengths pass; guiding the first and second light beams from the measurement object to the image sensor via the rotary filter in a time-series manner; adding, for each pixel, the outputs of the first light beams that have passed through the plurality of first filter regions and the outputs of the second light beams that have passed through the plurality of second filter regions; and measuring the temperature of the measurement object based on the output of the output adder or an averaging of the outputs of the output adder.

[0086] (Variation) Although the preferred embodiments of the present invention have been described above, the present invention is not limited to these embodiments and various modifications and changes are possible within the scope of the present invention. For example, as shown in Figure 9, a measuring terminal 41 (one or more) may be connected to a server 42 (functioning as the computer 7 in Figure 1) via the Internet.

[0087] Furthermore, in addition to the temperature measurement device and the temperature measurement method, the present invention can also be configured as a program for causing a computer to function as each means of the temperature measurement device described above. [Explanation of symbols]

[0088] 1 Objective lens, 2 Rotating filter, 3 Image sensor unit, 3a Image sensor, 3b Analog amplifier, 3c A / D conversion element, 4 Wavelength-specific image distribution circuit, 5 Image summation / non-summation processing circuit, 6 Memory bank, 7 Computer, 8 Trigger light source, 9 Trigger light receiving sensor, 10 Motor, 41 Measurement head, 42 Server

Claims

1. a single image sensor having two-dimensional pixels, each pixel being exposed simultaneously to light and outputting a digital signal; a rotary filter including a plurality of sets of first filter regions and second filter regions arranged in order, through which first light and second light having different wavelengths pass, respectively; a light guiding means for guiding the first light and the second light from the measurement object to the image sensor via the rotary filter in a time-series manner; an output summing unit that sums outputs of first light that has passed through a plurality of the first filter regions in units of each pixel, and that sums outputs of second light that has passed through a plurality of the second filter regions in units of each pixel; and a temperature measuring device for measuring the temperature of the object to be measured based on the output of the output adder or an averaging of the outputs of the output adder;

2. The temperature measuring device according to claim 1, characterized in that the temperature measuring device is used when the object to be measured is a low-temperature object not exceeding 500°C, and the temperature measuring device measures the temperature of the object to be measured based on the output of the output adder used to increase the output of the image sensor.

3. The temperature measuring device according to claim 1, characterized in that the temperature measuring device is used when the object to be measured is a high-temperature object exceeding 500°C, and the temperature of the object to be measured is measured based on the arithmetic average of the outputs of the output adder used to improve the accuracy of temperature measurement.

4. The temperature measuring device described in claim 1 is used when the object to be measured is a low-temperature object not exceeding 500°C, and when the object to be measured is a high-temperature object exceeding 500°C, the temperature of the object to be measured is measured based on the output of the output adder unit used to increase the output of the image sensor when the object to be measured is a low-temperature object not exceeding 500°C, and the temperature of the object to be measured is measured based on the arithmetic average of the outputs of the output adder unit used to increase the temperature measurement accuracy when the object to be measured is a high-temperature object exceeding 500°C.

5. The temperature measuring device described in claim 1, characterized in that it has a control unit that makes the rotation speed of the rotary filter slower than a reference value when the object to be measured is a low-temperature object not exceeding 500°C, and makes the rotation speed of the rotary filter faster than the reference value when the object to be measured is a high-temperature object exceeding 500°C.

6. 6. The temperature measuring device according to claim 5, wherein the control unit further sets the exposure time of the image sensor to be shorter than a predetermined value when the measurement target is a high-temperature object exceeding 500°C.

7. a step of providing a single image sensor having two-dimensional pixels, the image sensor being of a global shutter type in which each pixel is simultaneously exposed to light and outputs a digital signal, in an optical path from the measurement object; providing a rotary filter corresponding to the image sensor, the rotary filter including a plurality of sets of first filter regions and second filter regions arranged in order, through which first light and second light having different wavelengths pass, respectively; guiding the first light and the second light from the measurement object to the image sensor via the rotary filter in a time-series manner; adding, for each pixel, outputs of first light that has passed through a plurality of the first filter regions and adding, for each pixel, outputs of second light that has passed through a plurality of the second filter regions; measuring the temperature of the measurement object based on the output of the output adder or an averaging of the outputs of the output adder; A temperature measurement method comprising the steps of:

8. A program for causing a computer to function as each of the means of the temperature measuring device according to any one of claims 1 to 6.

Images