Non-contact temperature measuring device and non-contact temperature measuring method

Abstract

The non-contact temperature measuring device (1) includes a matching processing unit (24) that estimates pixel correspondence between images, a positional relationship estimation unit (25) that estimates a positional relationship between the infrared cameras (4A, 4B) and the object (A) based on the pixel correspondence, a luminance correction unit (26) that corrects the luminance corresponding to the temperature in the luminance image based on the positional relationship, and an emissivity correction unit (27) that estimates emissivity using the pixel correspondence and the luminance image and generates a temperature image based on the emissivity, where the matching processing unit (24) estimates pixel correspondence between images captured by the infrared cameras (4A, 4B) in the same light sensitivity wavelength band, and the emissivity correction unit (27) estimates emissivity using images captured by the infrared cameras (4A, 4B) in different light sensitivity wavelength bands.

Description

[Technical field]

[0001] The present disclosure relates to a non-contact temperature measurement device and a non-contact temperature measurement method. [Background technology]

[0002] An infrared camera captures an infrared image by detecting infrared light emitted by an object. An infrared image is an image showing a distribution of brightness according to the radiation intensity of infrared light emitted by an object, and the surface temperature of the object can be measured by using the infrared image. The relationship between the apparent radiation intensity of an object shown in the infrared image and the actual surface temperature of the object may vary depending on the infrared emissivity of the object and the positional relationship between the infrared camera and the object.

[0003] Here, the infrared emissivity of an object represents the ratio of the infrared energy radiated by the object according to Planck's law to the infrared energy radiated by a black body at the same temperature as the object, with the infrared energy radiated by the object being set at 1. Furthermore, the positional relationship between the infrared camera and the object includes, for example, the distance between the infrared camera and the object, and the angle of the object relative to the infrared camera.

[0004] In non-contact temperature measurement of an object using an infrared image, the temperature of the object may be erroneously measured due to changes in the relationship between the apparent radiation intensity of the object shown in the infrared image and the actual surface temperature of the object. In response to this, the infrared emissivity of the object can be estimated, and the temperature measurement error of the object can be corrected using this emissivity, and the positional relationship between the infrared camera and the object can be estimated, and the temperature measurement error of the object can be corrected using this positional relationship.

[0005] The radiation spectrum of infrared light emitted by an object has a different spectral shape depending on the temperature of the object. By utilizing this characteristic, it is possible to estimate the infrared emissivity of the object and estimate the positional relationship between the infrared camera and the object. For example, it is possible to specify an infrared sensitivity curve using multiple infrared images captured by multiple infrared cameras with different infrared light receiving sensitivity wavelength bands, and to estimate the emissivity of the object whose temperature is to be measured using the specified sensitivity curve, thereby correcting the effect of the emissivity on the measured temperature.

[0006] For example, Patent Document 1 describes a target detection device that estimates the positional relationship between an infrared camera and a target object. The target detection device described in Patent Document 1 performs binarization processing on infrared images of the same field of view captured by multiple infrared cameras with different infrared light receiving sensitivity wavelength bands, and extracts the target object by performing pixel matching between the binarized images. Then, it specifies the pixel correspondence between the binarized images in which the target object is captured, and estimates the positional relationship between the infrared camera and the object based on the specified pixel correspondence.

[0007] The invention described in Patent Document 1 is a target detection device that is mounted on a ship or an aircraft, etc., and captures and detects an object against the background of the sea surface or sky, etc., so the target object is captured as a point in the infrared image. Therefore, in a binary image in which a binarization process is performed on the distribution of pixel values ​​corresponding to the temperature of the object, for example, the luminance distribution, the target object is emphasized as a point light source with a higher temperature than the background, making it easier to extract the object from the binary image. [Prior art documents] [Patent documents]

[0008] [Patent Document 1] JP 2016-075615 A Summary of the Invention [Problem to be solved by the invention]

[0009] In non-contact temperature measurement of an object using an infrared camera, if it is possible to estimate the infrared emissivity of the object and the positional relationship between the infrared camera and the object using infrared images of the object whose temperature is to be measured taken by multiple infrared cameras with different infrared sensitivity wavelength bands, the influence these have on the measured temperature can be corrected.

[0010] Infrared images captured by a plurality of infrared cameras with different infrared light receiving sensitivity wavelength bands have significantly different distributions of pixel values ​​corresponding to temperature, for example, luminance distributions, between the images. In Patent Document 1, it is assumed that the target object is captured as a point in the infrared image, so the object is emphasized as a point light source in the binarized image obtained by binarizing the luminance distribution. Therefore, it is easy to identify a common object even among binarized images of images captured by multiple infrared cameras with different infrared light receiving sensitivity wavelength bands, and it is possible to estimate the positional relationship between the infrared camera and the object.

[0011] However, when multiple objects with various temperatures are captured in infrared images captured by multiple infrared cameras with different infrared light sensitivity wavelength bands, and the difference between the temperature of the object to be measured and the temperature of the background or other objects is not clear, the difference in brightness distribution corresponding to temperature is further emphasized between the binarized images. In this case, it becomes difficult to determine the pixel correspondence by pixel matching between the binarized images, making it impossible to extract the object from the image and to estimate the positional relationship between the infrared camera and the object. This causes a problem that it is not possible to correct the effect of the positional relationship between the infrared camera and the object on the measured temperature.

[0012] The present disclosure is intended to solve the above-mentioned problems, and aims to provide a non-contact temperature measurement device and a non-contact temperature measurement method that can correct the influence on the measured temperature of the infrared emissivity of an object and the positional relationship between the infrared camera and the object. [Means for solving the problem]

[0013] The non-contact temperature measuring device according to the present disclosure includes a matching processing unit that estimates pixel correspondence between multiple images captured by multiple infrared cameras of a field of view including a common area; a control unit that controls a change of the sensitivity wavelength band of at least one of the multiple infrared cameras; a positional relationship estimation unit that estimates a positional relationship between the infrared camera and an object whose temperature is to be measured that exists in the common area based on the estimated pixel correspondence; a pixel value correction unit that corrects pixel values ​​corresponding to temperature based on the estimated positional relationship between the infrared camera and the object; and the estimated pixel correspondence. By pixel value correction part Using the image with the pixel values ​​corrected, the infrared emissivity of the object is estimated, and based on the estimated emissivity, After being corrected by the pixel value correction unit The image capturing apparatus includes an emissivity correction unit that generates an image with corrected pixel values, and an infrared filter selection unit that selects a combination of light sensitivity wavelength bands to be used for measuring the temperature of an object using an image obtained by correcting pixel values ​​in an infrared image captured by the multiple infrared cameras of a field of view including a common area, the combinations of which include at least two of the combinations of light sensitivity wavelengths of the multiple infrared cameras being different, the pixel value correction unit and the emissivity correction unit being corrected. The control unit controls a change of the light sensitivity wavelength band of at least one of the multiple infrared cameras so that the light sensitivity wavelength bands of the multiple infrared cameras are the same or different. The matching processing unit estimates a correspondence relationship between pixels between images captured by the multiple infrared cameras of a field of view including a common area when the light sensitivity wavelength bands of at least two of the multiple infrared cameras are the same. The emissivity correction unit estimates the emissivity of infrared light of the object using an image captured by the multiple infrared cameras of a field of view including a common area when the light sensitivity wavelength bands of at least two of the multiple infrared cameras are different. Effect of the Invention

[0014] According to the present disclosure, a pixel correspondence is estimated between images captured by multiple infrared cameras using the same light sensitivity wavelength band to capture a field of view including a common area, a positional relationship between the infrared camera and an object whose temperature is to be measured that is located in the common area is estimated based on the estimated pixel correspondence, and the infrared emissivity of an object whose temperature is to be measured that is located in the common area is estimated using images captured by multiple infrared cameras using different light sensitivity wavelength bands to capture a field of view including the common area. Using an infrared image captured of a field of view including a common area, the positional relationship between the infrared camera and the object can be estimated, and the infrared emissivity of the object can be estimated.Therefore, the non-contact temperature measuring device of the present disclosure can correct the influence on the measured temperature of the infrared emissivity of the object and the positional relationship between the infrared camera and the object. [Brief description of the drawings]

[0015] [Figure 1] 1 is a block diagram showing an example of the configuration of a non-contact temperature measuring device according to a first embodiment. [Diagram 2] 4 is a flowchart showing the operation of the non-contact temperature measuring device according to the first embodiment. [Diagram 3] 11 is a flowchart showing a series of processes for shutter correction. [Figure 4] 11 is a flowchart showing a series of processes for stereo correction. [Diagram 5] 10 is a flowchart showing a series of processes for temperature correction. [Figure 6] 6A and 6B are block diagrams showing a hardware configuration for realizing the functions of the non-contact temperature measuring device according to the first embodiment. [Figure 7] FIG. 11 is a block diagram showing a configuration example of a non-contact temperature measuring device according to a second embodiment. [Figure 8] 10 is a flowchart showing the operation of the non-contact temperature measuring device according to the second embodiment. [Figure 9] 13 is a flowchart showing a series of processes for filter optimization and temperature correction. [Figure 10]1 is a graph showing peak wavelength characteristics of a blackbody radiation spectrum. [Figure 11] FIG. 11 is a block diagram showing a configuration example of a non-contact temperature measuring device according to a third embodiment. [Figure 12] 11 is a flowchart showing the operation of the non-contact temperature measuring device according to the third embodiment. [Figure 13] 10 is a flowchart showing a series of processes for temperature correction and temperature image synthesis. [Figure 14] FIG. 11 is a block diagram showing a configuration example of a non-contact temperature measuring device according to a fourth embodiment. [Figure 15] 10 is a flowchart showing the operation of the non-contact temperature measuring device according to the fourth embodiment. [Figure 16] 11 is a flowchart showing a series of processes for high-precision shutter correction. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0016] Embodiment 1 Fig. 1 is a block diagram showing a configuration example of a non-contact temperature measuring device 1 according to embodiment 1. In Fig. 1, the non-contact temperature measuring device 1 is a device that measures the temperature of an object A in a non-contact manner by using an infrared image of the object A whose temperature is to be measured, and includes a signal processing unit 2, a memory unit 3, an infrared camera 4A, an infrared camera 4B, an infrared filter unit 5A, an infrared filter changing unit 5, and a shutter unit 6.

[0017] The signal processing unit 2 performs temperature measurement processing of the object A using the infrared images captured by the infrared cameras 4A and 4B. The signal processing unit 2 includes a control unit 21, a sensitivity wavelength control unit 22, a luminance calibration unit 23, a matching processing unit 24, a positional relationship estimation unit 25, a luminance correction unit 26, and an emissivity correction unit 27. The infrared filter changing unit 5 is a device that includes an infrared filter unit 5B and an infrared filter unit 5C and has a function of switching the infrared filter unit.

[0018] The memory unit 3 is a storage device that stores information generated by signal processing by the luminance calibration unit 23, the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26 and the emissivity correction unit 27 provided in the signal processing unit 2, and outputs the information as required. Also, it is assumed that the memory unit 3 stores, for example, physical parameters indicating the physical characteristics of the infrared camera 4A and the infrared camera 4B as information at the time of shipment.

[0019] Infrared camera 4A and infrared camera 4B convert temperature changes caused by absorption of radiant energy of infrared light emitted from object A into electrical signals, and capture infrared images having pixel values ​​corresponding to the temperature of object A. Infrared camera 4A generates an infrared image capturing field of view B1, and infrared camera 4B generates an infrared image capturing field of view B2.

[0020] As shown in Fig. 1, the fields of view B1 and B2 include a common area, and the infrared cameras 4A and 4B are arranged so that the object A, the temperature of which is to be measured, is included in this common area. The infrared camera 4A detects the radiation light of the object A through the infrared filter unit 5A and generates an infrared image. The infrared camera 4B detects the radiation light of the object A through the infrared filter unit 5B or the infrared filter unit 5C and generates an infrared image. The infrared images generated by the infrared cameras 4A and 4B are output to the signal processing unit 2.

[0021] The light receiving sensitivity wavelength bands of infrared camera 4A and infrared camera 4B are, for example, a long wavelength band of 8 to 14 micrometers (μm), a medium wavelength band of 3 to 5 μm, or wavelength bands that differ from these wavelength bands by a certain wavelength, i.e., nearby wavelength bands. The light sensitivity wavelength bands of infrared cameras 4A and 4B may be the light sensitivity wavelength bands of the detection elements of the infrared cameras, or may be light sensitivity wavelength bands that have passed through an infrared filter section.

[0022] The peak wavelength of the radiation spectrum of infrared light from object A varies depending on the temperature. Generally, the radiation spectrum of an object near room temperature has a peak in the long wavelength band, but at high temperatures, for example, around 1000 K, it has a peak in the medium wavelength band. Therefore, infrared cameras 4A and 4B are used that are capable of capturing infrared images in a light receiving sensitivity wavelength band that has a peak at the expected temperature of the object to be measured.

[0023] The infrared camera 4A and the infrared camera 4B are configured to include a detection element that converts the emitted light into an electrical signal, an optical element that focuses the emitted light on the detection element, a readout circuit that reads out the electrical signal from the detection element, an AD conversion unit that converts the signal from the readout circuit into a digital signal, and a thermistor that measures the temperature of the camera itself. The detection element may be a bolometer type, a thermopile type, or a thermal diode type element. The focusing element may be a reflective optical element such as a mirror, or a refractive optical element such as a lens. The lens may be made of a material such as Si, Ge, or chalcogenide glass.

[0024] The infrared camera 4A and the infrared camera 4B may be attached to a common housing, or may be installed separately. In addition, although the non-contact temperature measuring device 1 equipped with two infrared cameras 4, i.e., the infrared camera 4A and the infrared camera 4B, is shown in Fig. 1, the non-contact temperature measuring device 1 may be equipped with three or more infrared cameras.

[0025] The infrared filter changing unit 5 changes the wavelength band of infrared light emitted from an object present in the field of view B2 and incident on the infrared camera 4B by switching between the infrared filter unit 5B and the infrared filter unit 5C. For example, the infrared filter changing unit 5 includes a filter wheel, and the infrared filter unit 5B and the infrared filter unit 5C inside the filter wheel, and the infrared filter unit 5B and the infrared filter unit 5C are switched by the rotation of the filter wheel.

[0026] The infrared filter unit 5A limits the wavelength band of infrared light emitted from an object present in the field of view B1 and incident on the infrared camera 4A. The infrared filter unit 5B and the infrared filter unit 5C limit the wavelength band of infrared light emitted from an object present in the field of view B2 and incident on the infrared camera 4B. For example, a bandpass filter formed by laminating multiple dielectric films is used as the infrared filter unit 5A, the infrared filter unit 5B, and the infrared filter unit 5C. The infrared filter unit may also be another optical element such as a mirror.

[0027] The infrared filter section 5A may have a structure that does not have a blocking band in the light-receiving sensitivity wavelength band of the infrared camera 4A and has a window or an opening. Moreover, the infrared filter changing section 5 may include three or more filters. The switching structure of the infrared filter section in the infrared filter changing section 5 is not limited to a filter wheel, and may be another wavelength band changing structure using a Fabry-Perot structure or the like.

[0028] The shutter unit 6 opens and closes the shutter in accordance with a control signal from the control unit 21, blocking the field of view B1 of the infrared camera 4A and blocking the field of view B2 of the infrared camera 4B. By opening and closing the shutter, the infrared image captured by the infrared camera 4A can be changed to a luminance image, and the infrared image captured by the infrared camera 4B can be changed to a luminance image. For example, the shutter unit 6 is made of a known material that has a high emissivity of infrared light.

[0029] The shutter unit 6 may be configured to include a temperature sensor. The temperature sensor measures the temperature of the shutter included in the shutter unit 6. For example, information indicating the temperature of the shutter measured by the temperature sensor is included in measurement-related information required for temperature measurement and is output to the signal processing unit 2.

[0030] The shutter unit 6 may be provided in each of the infrared cameras 4A and 4B. The shutters of the shutter unit 6 may also be disposed immediately before the infrared cameras 4A and 4B. This configuration has the advantage that it is not necessary to change the infrared filter unit when performing shutter correction, which will be described later.

[0031] Also, the shutter section 6 may be provided inside the camera housings of the infrared cameras 4A and 4B. For example, the shutter section 6 may be disposed between the optical element and the detection element. This has the advantage that the shutter section 6 can be manufactured integrally with the camera housing, making it possible to reduce the size of the shutter section 6 itself.

[0032] The shutter unit 6 may be configured integrally with the infrared filter changing unit 5, for example, a shielding plate may be disposed within the filter wheel. In this case, the number of moving parts can be reduced, which contributes to reducing the cost of the moving parts.

[0033] Control unit 21 performs control related to temperature measurement by acquiring infrared images and measurement-related information required for temperature measurement from infrared camera 4A and infrared camera 4B, and acquiring shutter-related information such as the measured temperature of the shutter from shutter unit 6. The infrared images from infrared camera 4A and infrared camera 4B include an infrared image of the shutter when the shutter provided in shutter unit 6 is closed, and further include measurement-related information required for measuring the temperature of the shutter.

[0034] For example, the control unit 21 outputs infrared images and measurement-related information acquired from the infrared cameras 4A and 4B to the brightness calibration unit 23, transmits control information for controlling the opening and closing of the shutter to the shutter unit 6, and transmits setting information indicating the light-receiving sensitivity wavelength band to the sensitivity wavelength control unit 22.

[0035] The sensitivity wavelength control unit 22 controls the switching of the infrared filter unit by the infrared filter changing unit 5 by transmitting a control signal to the infrared filter changing unit 5 that specifies the light receiving sensitivity wavelength band indicated by the setting information based on the setting information received from the control unit 21.

[0036] The luminance calibration unit 23 generates a calibration table using the infrared image of the shutter and measurement-related information necessary for measuring the temperature, which are input from the control unit 21. Then, the luminance calibration unit 23 performs luminance calibration of the infrared image of the object A, the temperature of which is to be measured, using the infrared image of the object A, the temperature of which is to be measured, and the calibration table read out from the memory unit 3.

[0037] The luminance calibration of an infrared image is a process of converting each pixel value of the infrared image into the radiance of an object, and as a result of the process, a luminance image is generated in which the pixel values ​​of the infrared image are converted into radiance. The calibration table has luminance calibration parameters set therein that are required when converting each pixel value of the infrared image into the radiance of an object. The luminance calibration parameters are data that indicate the correspondence between each pixel of the infrared image and the radiance of the object at the position corresponding to the pixel.

[0038] For example, the luminance calibration unit 23 performs shutter correction to generate the calibration table. The shutter correction is a process for calibrating the change in the correspondence between the radiance of infrared light from an object and the pixel value of an infrared image, which is caused by the change in the environmental temperature and the change in the temperature of the infrared camera itself. Specifically, the temperature of the shutter provided in the shutter unit 6 is assumed to be known, and the shutter is used as a reference target with known characteristics, and each pixel of the infrared image of the shutter and the radiance indicating the temperature of the shutter at the position corresponding to the pixel are identified to obtain the luminance calibration parameters. The luminance image generated by the luminance calibration is output from the luminance calibration unit 23 to the matching processing unit 24 and the luminance correction unit 26.

[0039] The matching processing unit 24 performs pixel matching processing on multiple luminance images of a common region (the same scene) acquired from the luminance calibration unit 23, and estimates the correspondence between pixels between the luminance images. For example, the luminance calibration unit 23 converts multiple infrared images of the same scene into multiple luminance images and outputs them to the matching processing unit 24. The matching processing unit 24 estimates the correspondence between pixels between the luminance images acquired from the luminance calibration unit 23.

[0040] The luminance image on which pixel matching processing is performed is an image based on infrared images of the same scene captured by the infrared cameras 4A and 4B in the same light receiving sensitivity wavelength band. The matching processing unit 24 estimates pixel matching information indicating the correspondence between pixels in the luminance image, and stores the estimated pixel matching information in the memory unit 3 and outputs it to the positional relationship estimation unit 25.

[0041] The positional relationship estimation unit 25 estimates the positional relationship between the infrared cameras 4A and 4B and the object A, the temperature of which is to be measured and exists in a common area, based on the pixel correspondence estimated by the matching processing unit 24. For example, the positional relationship estimation unit 25 reads out the pixel correspondence from the pixel matching information, and reads out information on the installation positions and orientation directions of the infrared cameras 4A and 4B from the measurement-related information from the memory unit 3. Next, the positional relationship estimation unit 25 uses the read-out information to calculate the distance between each pixel and the infrared cameras 4A and 4B, and generates a distance map including the distance to the infrared camera as a pixel value, and stores the generated distance map in the memory unit 3.

[0042] The luminance correction unit 26 is a pixel value correction unit that corrects pixel values ​​corresponding to temperature, based on the positional relationship between the infrared cameras 4A and 4B and the object A estimated by the positional relationship estimation unit 25. For example, luminance correction unit 26 inputs a luminance image from luminance calibration unit 23, and reads out the distance map estimated by positional relationship estimation unit 25 from memory unit 3. Then, luminance correction unit 26 corrects the radiance of each pixel of the luminance image using the distance map, and outputs the corrected luminance image to emissivity correction unit 27.

[0043] Even for objects having the same temperature and shape, the apparent radiance in the luminance image changes depending on the apparent size as seen by the infrared camera. Therefore, the apparent radiance corresponding to the temperature of an object that is located at the same distance from infrared camera 4A and the same distance from infrared camera 4B will differ depending on the apparent size as seen by infrared camera 4A and infrared camera 4B.

[0044] Furthermore, if an object has the same apparent size as seen by infrared camera 4A and infrared camera 4B, the apparent radiance in the luminance image changes according to the distance between the object and infrared camera 4A and infrared camera 4B. For example, a lookup table predicting the change in apparent radiance in the luminance image according to the distance between the infrared camera and the object and the apparent size of the object is created in advance and stored in memory unit 3. Luminance correction unit 26 calculates the true radiance for each pixel in the luminance image by correcting the apparent radiance using the lookup table read from memory unit 3.

[0045] When the size of object A, the temperature of which is to be measured, is known, the brightness correction unit 26 can calculate the true radiance for each pixel in the brightness image by referring to the lookup table using only the distance information indicating the distance between the infrared camera and the object. The luminance image in which the luminance of each pixel has been corrected is output from the luminance correcting section 26 to the emissivity correcting section 27 .

[0046] The emissivity correction unit 27 estimates the infrared emissivity of the object A using the pixel correspondence estimated by the matching processing unit 24 and the luminance image in which the luminance has been corrected by the luminance correction unit 26, and generates a temperature image in which pixel values ​​such as luminance have been corrected based on the estimated emissivity. For example, the emissivity correction unit 27 acquires a luminance image after luminance correction from the luminance correction unit 26, and reads out infrared light receiving sensitivity wavelength band information when the infrared camera 4A, the infrared camera 4B, the infrared filter unit 5A, the infrared filter unit 5B, and the infrared filter unit 5C are used from the memory unit 3. Then, the emissivity correction unit 27 estimates the emissivity of the object A using this information, and generates a temperature image by multiplying the luminance image after luminance correction by the estimated emissivity.

[0047] Although the signal processing unit 2 has been shown to have a configuration including the control unit 21, the sensitivity wavelength control unit 22, the luminance calibration unit 23, the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, and the emissivity correction unit 27, it is sufficient for the non-contact temperature measurement device 1 to include at least the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, and the emissivity correction unit 27. In this case, the control unit 21, the sensitivity wavelength control unit 22, and the luminance calibration unit 23 are provided in, for example, the infrared camera 4A or the infrared camera 4B.

[0048] Next, a specific operation will be described. FIG. 2 is a flowchart showing the operation of the non-contact temperature measuring device 1, illustrating a series of processes for temperature measurement by the non-contact temperature measuring device 1. First, the control unit 21 judges whether a certain time has passed since the previous shutter correction (step ST1). The shutter correction is a process for calibrating the change in the correspondence between the radiance of infrared light from an object and the pixel value of an infrared image, which is caused by the change in the environmental temperature and the temperature of the infrared camera itself. Here, since it is assumed that the environmental temperature and the temperature of the infrared camera itself do not change in a short time, the certain time is set to the time when at least one of them is expected to change. For example, it is a period of several hours to several days.

[0049] When the control unit 21 determines that a certain time has passed since the previous shutter correction (step ST1; YES), it outputs instruction information instructing the luminance calibration unit 23 to perform shutter correction. On the other hand, if the certain time has not passed since the previous shutter correction (step ST1; NO), the control unit 21 does not perform shutter correction, and outputs instruction information instructing the matching processing unit 24 to perform stereo correction via the luminance calibration unit 23.

[0050] The luminance calibration unit 23 performs shutter correction when a certain time has passed since the previous shutter correction (step ST2). For example, in accordance with the instruction information acquired from the control unit 21, the luminance calibration unit 23 identifies each pixel of an infrared image of the shutter (hereinafter referred to as a shutter image) and the radiance indicating the temperature of the shutter at a position corresponding to the pixel, with the temperature of the shutter being known, and calculates a luminance calibration parameter using the identified information. In addition, a conditional branch may be set in which the control unit 21 determines whether an event has occurred in which a change in temperature of the infrared camera itself and a change in the environmental temperature cause a change in the output pixels of the infrared camera, and if it determines that such an event has occurred, proceeds to shutter correction.

[0051] When the shutter correction is completed or a certain time has not elapsed since the previous shutter correction, the matching processing unit 24 and the positional relationship estimation unit 25 perform stereo correction (step ST3). The stereo correction is a process of performing pixel matching processing on a luminance image based on the infrared images captured by the infrared cameras 4A and 4B, and calculating the distance between the object A, the temperature measurement target, and the infrared cameras 4A and 4B for each pixel.

[0052] For example, the matching processing unit 24 acquires a plurality of luminance images from the luminance calibration unit 23, and performs pixel matching processing between the acquired luminance images to estimate the correspondence between pixels. The positional relationship estimation unit 25 calculates the distance between the object A, the temperature measurement target, and the infrared camera 4A and the infrared camera 4B for each pixel, based on the information indicating the pixel correspondence estimated by the matching processing unit 24. Although the case has been described in which the distance between the infrared camera and the object is calculated as the positional relationship between the infrared camera and the object, the angle of the object relative to the infrared camera may also be calculated.

[0053] Next, the luminance correction unit 26 and the emissivity correction unit 27 perform temperature correction on the luminance image for which the stereo correction has been completed (step ST4). The temperature correction is a process for correcting the influence of the distance between the object A and the infrared cameras 4A and 4B and the emissivity of infrared light of the object A on the luminance image based on the infrared images captured by the infrared cameras 4A and 4B.

[0054] For example, the luminance correction unit 26 reads out the distance map estimated by the positional relationship estimation unit 25 from the memory unit 3, and uses the distance map to correct the radiance of each pixel of the luminance image input from the luminance calibration unit 23. The emissivity correction unit 27 acquires the luminance image after luminance correction acquired from the luminance correction unit 26, and reads out the sensitivity wavelength band information when the infrared camera 4A, the infrared camera 4B, the infrared filter unit 5A, the infrared filter unit 5B, and the infrared filter unit 5C are used from the memory unit 3. Next, the emissivity correction unit 27 estimates the emissivity of the object A using this information, and generates a temperature image by multiplying the luminance image after luminance correction by the estimated emissivity.

[0055] The emissivity correction unit 27 outputs the temperature image finally determined by the brightness correction unit 26 to an external device not shown in Fig. 1 (step ST5). Here, the external device is a device capable of communicating with the non-contact temperature measuring device 1, and may be, for example, a display device capable of displaying a temperature image. Note that the emissivity correction unit 27 may output distance information and emissivity information secondarily generated in the stereo correction or temperature correction together with the temperature image.

[0056] Next, the process of each step in FIG. 2 will be described in detail. FIG. 3 is a flowchart showing a series of processes for shutter correction, and shows a series of processes in step ST2 of FIG. First, the control unit 21 transmits a control signal for closing the shutters to the shutter unit 6. The shutter unit 6 closes the shutters in accordance with the control signal, blocking the fields of view of the infrared cameras 4A and 4B (step ST1-1).

[0057] Next, the infrared cameras 4A and 4B capture images of the shutters blocking the visual fields B1 and B2 to generate shutter images. The control unit 21 acquires the shutter images from the infrared cameras 4A and 4B (step ST1-2).

[0058] The brightness calibration unit 23 acquires from the control unit 21, in accordance with instruction information instructing shutter correction, the shutter image, the temperature of the shutter unit 6, and infrared images of object A captured by the infrared cameras 4A and 4B, and reads from the memory unit 3 sensitivity wavelength band information indicating the light receiving sensitivity wavelength bands of the infrared cameras 4A and 4B, transmission wavelength band information indicating the transmission wavelength bands of the infrared filter units currently set therefor, and the like. Then, using the information acquired from the control unit 21 and the information read from the memory unit 3, the brightness calibration unit 23 calculates brightness calibration parameters required for converting the pixel values ​​of the infrared image into the radiance of object A for each pixel of the infrared image (step ST1-3). The luminance calibration unit 23 generates a calibration table in which parameters for luminance calibration for each pixel are set, and stores the generated calibration table in the memory unit 3.

[0059] Next, the control unit 21 outputs the sensitivity wavelength band information to the sensitivity wavelength control unit 22 to specify the light receiving sensitivity wavelength band. The sensitivity wavelength control unit 22 outputs a control signal to the infrared filter changing unit 5 to switch to an infrared filter unit that transmits the wavelength band indicated by the sensitivity wavelength band information acquired from the control unit 21. The infrared filter changing unit 5 changes the infrared filter unit set in the field of view B2 of the infrared camera 4B to the infrared filter unit 5B or the infrared filter unit 5C in accordance with the control signal (step ST1-4).

[0060] After changing the infrared filter unit, the control unit 21 returns to the process of step ST1-2 and performs the processes of steps ST1-2 and ST1-3 again. This series of processes is performed for all the infrared filter units included in the infrared filter changing unit 5. In addition, when the shutter section 6 is provided between the infrared filter changing section 5 and the infrared camera 4A and the infrared camera 4B, these processes may be omitted.

[0061] When the above-mentioned processing is completed, the control unit 21 transmits a control signal for opening the shutter to the shutter unit 6. The shutter unit 6 opens the shutter in accordance with the control signal, and ends blocking the fields of view of the infrared cameras 4A and 4B (step ST1-5).

[0062] FIG. 4 is a flowchart showing a series of stereo correction processes, showing a series of processes in step ST3 of FIG. First, the control unit 21 specifies a light-receiving sensitivity wavelength band by outputting sensitivity wavelength band information to the sensitivity wavelength control unit 22. It is assumed that the infrared cameras 4A and 4B have the same light-receiving sensitivity wavelength band, and the infrared filter units 5A and 5B have the same transmission wavelength band. The control unit 21 outputs sensitivity wavelength band information that specifies the transmission wavelength band of the infrared filter unit 5B as the sensitivity wavelength to the infrared filter changing unit 5 so that the infrared cameras 4A and 4B have the same light receiving sensitivity wavelength band. The sensitivity wavelength control unit 22 outputs a control signal for switching the infrared filter unit to the infrared filter changing unit 5 based on the sensitivity wavelength band information acquired from the control unit 21. The infrared filter changing unit 5 changes the infrared filter unit set in the field of view B2 of the infrared camera 4B to the infrared filter unit 5B in accordance with the control signal (step ST2-1).

[0063] With the infrared filter unit 5A set in the infrared camera 4A and the infrared filter unit 5B set in the infrared camera 4B, the infrared camera 4A captures an image of the field of view B1, and the infrared camera 4B captures an image of the field of view B2. The control unit 21 acquires the infrared images captured by the infrared cameras 4A and 4B (step ST2-2). The infrared images acquired by the control unit 21 are images captured by the infrared cameras 4A and 4B in the same light-receiving sensitivity wavelength band (hereinafter referred to as single-band images).

[0064] The control unit 21 outputs the infrared images acquired from the infrared cameras 4A and 4B and instruction information for performing luminance calibration to convert these infrared images into luminance images to the luminance calibration unit 23. The luminance calibration unit 23 reads out a calibration table from the memory unit 3 in accordance with the instruction information, and converts the pixel values ​​of the infrared images into radiance using luminance calibration parameters corresponding to the same light-receiving sensitivity wavelength band included in the calibration table, thereby generating a luminance image (step ST2-3). The luminance calibration unit 23 outputs the generated luminance image to the matching processing unit 24.

[0065] The matching processing unit 24 calculates the correspondence between pixels between the luminance images by performing pixel matching processing between a luminance image based on an infrared image captured by the infrared camera 4A and a luminance image based on an infrared image captured by the infrared camera 4B (step ST2-4). The distribution of infrared radiance is also the same between luminance images based on infrared images captured in the same light-receiving sensitivity wavelength band by infrared camera 4A and infrared camera 4B, and therefore the luminance distribution caused by the temperature of the same object A is also the same. This makes it easy to identify the object to be matched, and makes it easy to perform pixel matching between luminance images. The matching processing unit 24 stores in the memory unit 3 information indicating the pixel correspondence obtained by pixel matching.

[0066] The positional relationship estimation unit 25 acquires information indicating the pixel correspondence between the luminance images estimated by the matching processing unit 24, and information indicating the installation positions and orientation directions of the infrared cameras 4A and 4B that have been stored in advance in the memory unit 3. Next, the positional relationship estimation unit 25 performs general stereo processing by calculation based on geometric optics using the acquired information to estimate the distance between the object A and the infrared cameras 4A and 4B in the luminance image for each pixel (step ST2-5). A distance map indicating the estimated distance for each pixel is stored in the memory unit 3.

[0067] FIG. 5 is a flowchart showing a series of processes for temperature correction, which shows a series of processes in step ST4 of FIG. First, the control unit 21 specifies a light-receiving sensitivity wavelength band by outputting sensitivity wavelength band information to the sensitivity wavelength control unit 22. The infrared cameras 4A and 4B have the same light-receiving sensitivity wavelength band, and the infrared filter unit 5A and the infrared filter unit 5C have different transmission wavelength bands. The control unit 21 outputs sensitivity wavelength band information that specifies the transmission wavelength band of the infrared filter unit 5C as the sensitivity wavelength to the infrared filter changing unit 5 so that the infrared cameras 4A and 4B have different light receiving sensitivity wavelength bands. The sensitivity wavelength control unit 22 outputs a control signal for switching the infrared filter unit to the infrared filter changing unit 5 based on the sensitivity wavelength band information acquired from the control unit 21. The infrared filter changing unit 5 changes the infrared filter unit set in the field of view B2 of the infrared camera 4B to the infrared filter unit 5C in accordance with the control signal (step ST3-1).

[0068] With the infrared filter unit 5A set in the infrared camera 4A and the infrared filter unit 5C set in the infrared camera 4B, the infrared camera 4A captures an image of the field of view B1, and the infrared camera 4B captures an image of the field of view B2. The control unit 21 acquires the infrared images captured by the infrared cameras 4A and 4B (step ST3-2). The infrared images acquired by the control unit 21 are images captured by the infrared cameras 4A and 4B in different light receiving sensitivity wavelength bands (hereinafter referred to as multiband images).

[0069] The control unit 21 outputs the infrared images acquired from the infrared cameras 4A and 4B and instruction information for performing luminance calibration to convert these infrared images into luminance images to the luminance calibration unit 23. The luminance calibration unit 23 reads a calibration table from the memory unit 3 according to the instruction information, and converts the pixel values ​​of the infrared images into radiance using luminance calibration parameters corresponding to different light receiving sensitivity wavelength bands included in the calibration table, thereby generating a luminance image (step ST3-3). The luminance calibration unit 23 outputs the generated luminance image to the luminance correction unit 26.

[0070] Next, the luminance correction unit 26 corrects the radiance of each pixel of the luminance image acquired from the luminance calibration unit 23 using the distance map read from the memory unit 3 (step ST3-4). As described above, the apparent radiance corresponding to the temperature of the object A existing at a position where the distance between the object A and the infrared camera 4A is the same as the distance between the object A and the infrared camera 4B differs depending on the apparent size of the object A as seen from the infrared camera 4A and the infrared camera 4B. For example, a lookup table predicting a change in apparent radiance in a luminance image according to the distance between the infrared camera and object A and the apparent size of object A is created in advance and stored in memory unit 3. Luminance correction unit 26 corrects the apparent radiance of object A using the lookup table read from memory unit 3, thereby calculating the true radiance for each pixel in the luminance image.

[0071] When the size of object A, the temperature of which is to be measured, is known, the brightness correction unit 26 can calculate the true radiance for each pixel in the brightness image by referring to the lookup table using only the distance information indicating the distance between the infrared camera and the object. The luminance image in which the luminance of each pixel has been corrected is output from the luminance correcting section 26 to the emissivity correcting section 27 .

[0072] Moreover, deconvolution can be used for brightness correction. If the spread distribution of the point images of the optical elements mounted on each of the infrared cameras 4A and 4B is known, the true radiance distribution can be estimated by applying deconvolution to the apparent radiance distribution in the luminance image. By using the distance information, it is possible to specify a change in the spread distribution of the point image according to a change in the distance between the object A and the infrared camera 4A and infrared camera 4B.

[0073] Next, the emissivity correction unit 27 estimates the infrared emissivity of the object A, and multiplies the luminance image by the estimated emissivity to generate a temperature image, which is a luminance image that has been subjected to luminance correction using the emissivity (step ST3-5). For example, even if two objects have the same radiance, if the objects have different emissivities, they will have different surface temperatures. Therefore, in order to measure the correct surface temperature, it is necessary to correct the radiance using the emissivity.

[0074] Therefore, the emissivity correction unit 27 estimates the infrared emissivity of the object A using the luminance image after luminance correction acquired from the luminance correction unit 26, the pixel correspondence between the luminance images read from the memory unit 3, and sensitivity wavelength band information indicating the light receiving sensitivity wavelength bands of the infrared cameras 4A and 4B. Then, the emissivity correction unit 27 multiplies the luminance image after luminance correction by the luminance correction unit 26 by the estimated emissivity.

[0075] The luminance correction using emissivity is a process based on the following formula (1). In the following formula (1), L O is the spectral radiance of object A, and L B is the blackbody spectral radiance, and T o is the surface temperature of object A, and ρ o is the infrared emissivity of object A. L o =ρ o L B (T o ,λ) (1)

[0076] The true radiances L1 and L2 calculated in step ST3-4 from the luminance images based on the infrared images captured by the infrared cameras 4A and 4B, respectively, are expressed by the following formulas (2) and (3). In the following formulas (2) and (3), η1(λ) is a sensitivity curve taking into account the transmission wavelength band of the infrared filter unit 5A set in the infrared camera 4A. η1(λ) is a sensitivity wavelength band taking into account the transmission wavelength band of the infrared filter unit 5C set in the infrared camera 4B. L1 = ∫η1(λ)L o dλ=ρ o ∫η1(λ)L B (T o ,λ)dλ (2) L2=∫η2(λ)L o dλ=ρ o ∫η2(λ)L B (T o ,λ)dλ (3)

[0077] In the above formula (2) and the above formula (3), To and ρ o are unknown variables. Two equations can be formulated for these unknown variables, and by solving these as simultaneous equations, To and ρ o By performing this process for all pixels in the brightness image, a temperature image can be generated. In addition, by performing the series of processes shown in Figure 2, the brightness is corrected using the distance between the infrared camera and object A and the emissivity of object A, making it possible to accurately measure the surface temperature of object A.

[0078] A feature of temperature measurement by the non-contact temperature measuring device 1 is that the light receiving sensitivity wavelength bands of the infrared cameras 4A and 4B are changed. In order to estimate the infrared emissivity of object A, it is necessary for infrared camera 4A and infrared camera 4B to have different light receiving sensitivity wavelength bands. On the other hand, luminance images based on infrared images captured in different light-receiving wavelength bands have significantly different luminance values ​​and their distributions, making pixel matching difficult and making it difficult to estimate distances based on the correspondence between pixels.

[0079] Therefore, the non-contact temperature measuring device 1 changes the light receiving sensitivity wavelength bands of the infrared cameras 4A and 4B in the distance correction in step ST2 and the temperature correction in step ST3. For example, in step ST2, by using infrared images captured in the same light-receiving sensitivity wavelength band, pixel matching can be easily performed between luminance images. On the other hand, in step ST3, the emissivity can be estimated using infrared images captured in different light-receiving sensitivity wavelength bands, and luminance correction can be performed using the emissivity. This allows the non-contact temperature measuring device 1 to correct the influence of the infrared emissivity of the object A and the positional relationship between the object A and the infrared cameras 4A and 4B, and the infrared camera 4A.

[0080] The non-contact temperature measuring device 1 uses luminance images based on infrared images captured in the same light-receiving sensitivity wavelength band, making it easy to identify corresponding luminance distributions between luminance images, and making it possible to omit techniques such as binarization of luminance images or extraction of contours (common shapes) from luminance images in pixel matching processing.

[0081] Next, a hardware configuration for realizing the functions of the non-contact temperature measuring device 1 will be described. The functions of the control unit 21, sensitivity wavelength control unit 22, luminance calibration unit 23, matching processing unit 24, positional relationship estimation unit 25, luminance correction unit 26, and emissivity correction unit 27 included in the non-contact temperature measurement device 1 are realized by processing circuits. That is, the non-contact temperature measurement device 1 includes a processing circuit for executing the processes from step ST2-1 to step ST2-5 shown in Fig. 4 and the processes from step ST3-1 to step ST3-5 shown in Fig. 5. The processing circuit may be dedicated hardware, or may be a CPU (Central Processing Unit) that executes a program stored in a memory.

[0082] Fig. 6A is a block diagram showing a hardware configuration for realizing the functions of the non-contact temperature measuring device 1. Fig. 6B is a block diagram showing a hardware configuration for executing software for realizing the functions of the non-contact temperature measuring device 1. In Figs. 6A and 6B, an input interface 100 is an interface for relaying various data stored in the memory unit 3 and relaying infrared images acquired from the infrared cameras 4A and 4B. An output interface 101 is an interface for relaying a control signal output from the sensitivity wavelength control unit 22 to the infrared filter changing unit 5 and temperature image data output from the emissivity correction unit 27.

[0083] When the processing circuit is the dedicated hardware processing circuit 102 shown in FIG. 6A, the processing circuit 102 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a combination thereof. The functions of the control unit 21, the sensitivity wavelength control unit 22, the luminance calibration unit 23, the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26 and the emissivity correction unit 27 included in the non-contact temperature measurement device 1 may be realized by separate processing circuits, or these functions may be realized together by a single processing circuit.

[0084] 6B, the functions of the control unit 21, sensitivity wavelength control unit 22, luminance calibration unit 23, matching processing unit 24, positional relationship estimation unit 25, luminance correction unit 26, and emissivity correction unit 27 included in the non-contact temperature measurement device 1 are realized by software, firmware, or a combination of software and firmware. The software or firmware is written as a program and stored in the memory 104.

[0085] The processor 103 reads out and executes the programs stored in the memory 104, thereby realizing the functions of the control unit 21, the sensitivity wavelength control unit 22, the luminance calibration unit 23, the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26 and the emissivity correction unit 27 provided in the non-contact temperature measuring device 1. For example, the non-contact temperature measuring device 1 includes a memory 104 for storing a program that, when executed by the processor 103, results in the processing of steps ST2-1 to ST2-5 shown in FIG. 4 and the processing of steps ST3-1 to ST3-5 shown in FIG. 5 being executed. These programs cause the computer to execute the procedures or methods of processing performed by the control unit 21, the sensitivity wavelength control unit 22, the luminance calibration unit 23, the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, and the emissivity correction unit 27. The memory 104 may be a computer-readable storage medium storing programs for causing the computer to function as the control unit 21, the sensitivity wavelength control unit 22, the luminance calibration unit 23, the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, and the emissivity correction unit 27.

[0086] When the processor 103 inputs an infrared image, it stores data obtained by temperature measurement using the infrared image in memory 104, and stores a temperature image obtained by performing various corrections on the data stored in memory 104 in memory 104. Furthermore, the processor 103 reads out the temperature image stored in memory 104, and outputs it to an external device.

[0087] Memory 104 is memory unit 3 shown in FIG. 1, and corresponds to, for example, a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable Read Only Memory), an EEPROM (Electrically-EPROM) (registered trademark), etc.

[0088] Some of the functions of the control unit 21, the sensitivity wavelength control unit 22, the luminance calibration unit 23, the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26 and the emissivity correction unit 27 included in the non-contact temperature measuring device 1 may be realized by dedicated hardware, and the remaining functions may be realized by software or firmware. For example, the functions of the control unit 21, the sensitivity wavelength control unit 22, and the luminance calibration unit 23 may be realized by a processing circuit 102 which is dedicated hardware, and each function of the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, and the emissivity correction unit 27 may be realized by a processor 103 reading and executing a program stored in a memory 104. In this way, the processing circuit can realize the above functions by hardware, software, firmware, or a combination of these.

[0089] As described above, the non-contact temperature measuring device 1 according to the first embodiment includes a matching processing unit 24 which estimates pixel correspondence between a plurality of images captured by the infrared cameras 4A and 4B of a field of view including the same scene, a positional relationship estimation unit 25 which estimates the positional relationship between the infrared cameras 4A and 4B and the object A based on the pixel correspondence, a luminance correction unit 26 which corrects the luminance corresponding to the temperature based on the positional relationship between the infrared cameras 4A and 4B and the object A, and an emissivity correction unit 27 which estimates the infrared emissivity of the object A using the estimated pixel correspondence and the image with the luminance corrected, and generates an image with the luminance corrected based on the estimated emissivity. The matching processing unit 24 estimates the pixel correspondence between the images captured by the infrared camera 4A and the infrared camera 4B in the same light sensitivity wavelength band, and the emissivity correction unit 27 estimates the infrared light emissivity of the object A using the images captured by the infrared camera 4A and the infrared camera 4B in different light sensitivity wavelength bands. In the non-contact temperature measuring device 1, the positional relationship between the infrared cameras 4A and 4B and the object A can be estimated using an infrared image captured of a field of view including the same scene, and the infrared emissivity of the object A can be estimated. This allows the non-contact temperature measuring device 1 to correct the influence on the measured temperature of the infrared emissivity of the object A and the positional relationship between the infrared cameras 4A and 4B and the object A.

[0090] The non-contact temperature measuring device 1 according to the first embodiment includes a shutter unit 6 that blocks the field of view of the infrared camera 4A and the infrared camera 4B, an infrared filter changing unit 5 that has infrared filter units 5A to 5C that pass infrared light of different wavelength bands and switches between the infrared filter units 5A to 5C that pass infrared light incident on at least one of the infrared camera 4A and the infrared camera 4B, a sensitivity wavelength control unit 22 that controls the switching of the infrared filter units 5A to 5C, and a brightness calibration unit 23 that converts a plurality of infrared images captured by the infrared camera 4A and the infrared camera 4B into a brightness image that shows a distribution of brightness values ​​for each pixel corresponding to temperature. A matching processing unit 24 estimates the correspondence relationship between pixels between the brightness images. With this configuration, the non-contact temperature measuring device 1 can estimate the positional relationship between the infrared cameras 4A and 4B and the object A using infrared images captured of a field of view including the same scene, and can estimate the infrared light emissivity of the object A.

[0091] In the non-contact temperature measuring device 1 according to the first embodiment, the light receiving sensitivity wavelength bands of the infrared cameras 4A and 4B are a wavelength band from 8 to 14 μm, a wavelength band from 3 to 5 μm, and a wavelength band that differs from these wavelength bands by a certain wavelength. This allows the non-contact temperature measuring device 1 to measure various temperature ranges.

[0092] In the non-contact temperature measuring device 1 according to the first embodiment, the sensitivity wavelength control unit 22 controls the switching of the infrared filter units 5A to 5C to match the light sensitivity wavelength bands of at least two of the infrared cameras 4A and 4B. This enables the non-contact temperature measuring device 1 to estimate the correspondence between pixels between images captured in the same light sensitivity wavelength band.

[0093] In the non-contact temperature measuring device 1 according to the first embodiment, the shutter unit 6 is configured integrally with the infrared filter changing unit 5, and blocks the field of view of the infrared camera into which infrared light is incident that has passed through the infrared filter units 5A to 5C switched by the infrared filter changing unit 5. This enables the brightness calibrator 23 to perform shutter correction and generate a calibration table.

[0094] In the non-contact temperature measuring device 1 according to the first embodiment, the sensitivity wavelength control unit 22 can change the number of combinations of the light receiving sensitivity wavelength bands of the infrared camera 4A and the infrared camera 4B to at least three or more. This allows the non-contact temperature measuring device 1 to measure various temperature ranges.

[0095] In the non-contact temperature measurement method according to the first embodiment, the matching processing unit 24 estimates pixel correspondence between a plurality of images captured by the infrared cameras 4A and 4B of a field of view including the same scene, the positional relationship estimation unit 25 estimates the positional relationship between the infrared cameras 4A and 4B and the object A based on the pixel correspondence, the luminance correction unit 26 corrects the luminance corresponding to the temperature based on the positional relationship between the infrared cameras 4A and 4B and the object A, and the emissivity correction unit 27 estimates the infrared emissivity of the object A using the estimated pixel correspondence and the images with the luminance corrected, and generates an image with the luminance corrected based on the estimated emissivity. The matching processing unit 24 estimates the pixel correspondence between the images captured by the infrared cameras 4A and 4B in the same light sensitivity wavelength band, and the emissivity correction unit 27 estimates the infrared emissivity of the object A using the images captured by the infrared cameras 4A and 4B in different light sensitivity wavelength bands. By executing this method, the non-contact temperature measuring device 1 can correct the influence on the measured temperature of the infrared emissivity of the object A and the positional relationship between the infrared cameras 4A and 4B and the object A.

[0096] Embodiment 2 Fig. 7 is a block diagram showing a configuration example of a non-contact temperature measuring device 1A according to embodiment 2. In Fig. 7, the non-contact temperature measuring device 1A is a device that measures the temperature of an object A in a non-contact manner by using an infrared image of the object A whose temperature is to be measured, and includes a signal processing unit 2A, a memory unit 3, an infrared camera 4A, an infrared camera 4B, an infrared filter unit 5A, an infrared filter changing unit 5-1, and a shutter unit 6.

[0097] The signal processing unit 2A performs temperature measurement processing of the object A using the infrared images captured by the infrared cameras 4A and 4B. The signal processing unit 2A includes a control unit 21, a sensitivity wavelength control unit 22, a luminance calibration unit 23, a matching processing unit 24, a positional relationship estimation unit 25, a luminance correction unit 26, an emissivity correction unit 27, and an infrared filter selection unit 28. The infrared filter changing unit 5-1 is a device that includes an infrared filter unit 5B, an infrared filter unit 5C, and an infrared filter unit 5D and has a function of switching the infrared filter unit.

[0098] The infrared filter unit 5A limits the wavelength band of infrared light emitted from an object present in the field of view B1 and incident on the infrared camera 4A. The infrared filter unit 5B, the infrared filter unit 5C, and the infrared filter unit 5D limit the wavelength band of infrared light emitted from an object present in the field of view B2 and incident on the infrared camera 4B. For example, a bandpass filter formed by laminating multiple dielectric films is used as the infrared filter unit 5A, the infrared filter unit 5B, the infrared filter unit 5C, and the infrared filter unit 5D. The infrared filter unit may also be another optical element such as a mirror.

[0099] For example, the transmission wavelength bandwidth of infrared filter portion 5A, infrared filter portion 5B, infrared filter portion 5C, and infrared filter portion 5D is 1 μm. Also, as an example, the center wavelength of the transmission wavelength band is 9 μm for infrared filter portion 5A and infrared filter portion 5B, 8 μm for infrared filter portion 5C, and 10 μm for infrared filter portion 5D.

[0100] The infrared filter changing unit 5-1 changes the wavelength band of infrared light emitted from an object present in the field of view B2 and incident on the infrared camera 4B by switching between the infrared filter unit 5B, the infrared filter unit 5C, and the infrared filter unit 5D. For example, the infrared filter changing unit 5-1 includes a filter wheel, and the infrared filter unit 5B, the infrared filter unit 5C, and the infrared filter unit 5D within the filter wheel, and the infrared filter unit 5B, the infrared filter unit 5C, and the infrared filter unit 5D are switched between by the rotation of the filter wheel.

[0101] At least one of the infrared filter section 5B, the infrared filter section 5C, and the infrared filter section 5D included in the infrared filter changing section 5-1 has the same transmission wavelength band as the infrared filter section 5A. In addition, at least one of the infrared filter section 5B, the infrared filter section 5C, and the infrared filter section 5D may have a structure having a window or an opening that does not have a blocking band in the light receiving sensitivity wavelength bands of the infrared cameras 4A and 4B. Moreover, the infrared filter changing section 5-1 may include three or more filters. The switching structure of the infrared filter section in the infrared filter changing section 5-1 is not limited to a filter wheel, and may be another wavelength band changing structure using a Fabry-Perot structure or the like.

[0102] The combination of infrared light transmission wavelength bands set by changing the infrared filter section by the infrared filter changing section 5-1 must be at least three types. For example, in the non-contact temperature measuring device 1A, the central wavelength of the light sensitivity wavelength band of the infrared camera 4A and the central wavelength of the light sensitivity wavelength band of the infrared camera 4B can be set to three different combinations: 9 μm-9 μm, 9 μm-8 μm, and 9 μm-10 μm.

[0103] The central wavelength or transmission bandwidth of the infrared filter section included in the infrared filter changing section 5-1 may not be as described above, and the number of infrared filter sections may be changed. For example, the number of infrared filter sections included in the infrared filter changing section 5-1 may be two, or another infrared filter changing section including two or more infrared filter sections may be newly added to the infrared filter changing section 5-1.

[0104] The infrared filter selection unit 28 selects a combination of light sensitivity wavelength bands to be used for measuring the temperature of the object A using a luminance image based on an infrared image captured by the infrared camera 4A and the infrared camera 4B, respectively, of a field of view including a common area using a combination of two or more light sensitivity wavelength bands, the luminance image having the radiance of each pixel corrected by the luminance correction unit 26 and the emissivity correction unit 27.

[0105] For example, the infrared filter selection unit 28 judges whether or not the current combination of infrared filter units is an optimal combination that can improve the accuracy of the measured temperature of the object A, based on the luminance image acquired from the emissivity correction unit 27. If it is determined that the combination of infrared filter units is not optimal, the infrared filter selection unit 28 outputs instruction information to change the infrared filter units to the control unit 21, and if it is determined that the combination is optimal, it outputs the luminance image acquired from the emissivity correction unit 27 as it is as a temperature image.

[0106] Although the signal processing unit 2A has been shown to include the control unit 21, the sensitivity wavelength control unit 22, the luminance calibration unit 23, the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, the emissivity correction unit 27, and the infrared filter selection unit 28, it is sufficient for the non-contact temperature measurement device 1A to include at least the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, the emissivity correction unit 27, and the infrared filter selection unit 28. In this case, the control unit 21, the sensitivity wavelength control unit 22, and the luminance calibration unit 23 are provided in, for example, the infrared camera 4A or the infrared camera 4B.

[0107] Next, a specific operation will be described. Fig. 8 is a flow chart showing the operation of the non-contact temperature measuring device 1A, and shows a series of processes for temperature measurement by the non-contact temperature measuring device 1A. In Fig. 8, the processes of steps ST1, ST2, ST3, and ST5 are similar to the processes of steps ST1, ST2, ST3, and ST5 in Fig. 2, and therefore description thereof will be omitted.

[0108] The luminance correction unit 26, the emissivity correction unit 27, and the infrared filter selection unit 28 perform filter optimization and temperature correction on the luminance image for which the stereo correction has been completed (step ST4A). Filter optimization is a process of selecting an optimal combination of infrared filters that can improve the accuracy of the measured temperature of object A. Temperature correction is a process of correcting the influence of the distance between object A and infrared cameras 4A and 4B, and the infrared emissivity of object A, on a luminance image based on the infrared images captured by infrared cameras 4A and 4B.

[0109] FIG. 9 is a flowchart showing a series of processes for filter optimization and temperature correction, and shows a series of processes in step ST4A of FIG. First, the control unit 21 specifies a light-receiving sensitivity wavelength band by outputting sensitivity wavelength band information to the sensitivity wavelength control unit 22. The infrared cameras 4A and 4B have the same light-receiving sensitivity wavelength band, the infrared filter units 5A and 5B have the same transmission wavelength band (9 μm-9 μm), the infrared filter units 5A and 5C have different transmission wavelength bands (9 μm-8 μm), and the infrared filter units 5A and 5D have different transmission wavelength bands (9 μm-10 μm).

[0110] The control unit 21 outputs sensitivity wavelength band information to the infrared filter changing unit 5-1, which specifies the transmission wavelength band of the infrared filter unit 5C or the infrared filter unit 5D as the sensitivity wavelength so that the infrared cameras 4A and 4B have different light receiving sensitivity wavelength bands. The sensitivity wavelength control unit 22 outputs a control signal for switching the infrared filter unit to the infrared filter changing unit 5 based on the sensitivity wavelength band information acquired from the control unit 21. In accordance with the control signal, the infrared filter changing unit 5 changes the infrared filter unit to be set in the field of view B2 of the infrared camera 4B to the infrared filter unit 5C or the infrared filter unit 5D (step ST4A-1).

[0111] With the infrared filter unit 5A set in the infrared camera 4A and the infrared filter unit 5C or 5D set in the infrared camera 4B, the infrared camera 4A captures an image of the field of view B1, and the infrared camera 4B captures an image of the field of view B2. The control unit 21 acquires the infrared images captured by the infrared cameras 4A and 4B (step ST4A-2). The infrared image acquired by the control unit 21 is a multiband image captured by the infrared cameras 4A and 4B in different light receiving sensitivity wavelength bands.

[0112] The control unit 21 outputs the infrared images acquired from the infrared cameras 4A and 4B and instruction information for performing luminance calibration to convert these infrared images into luminance images to the luminance calibration unit 23. The luminance calibration unit 23 reads a calibration table from the memory unit 3 in accordance with the instruction information, and converts the pixel values ​​of the infrared images into radiance using luminance calibration parameters corresponding to different light-receiving sensitivity wavelength bands included in the calibration table, thereby generating a luminance image (step ST4A-3). The luminance calibration unit 23 outputs the generated luminance image to the luminance correction unit 26.

[0113] The luminance correction unit 26 uses the distance map read from the memory unit 3 to correct the radiance of each pixel of the luminance image acquired from the luminance calibration unit 23 (step ST4A-4). For example, the luminance correction unit 26 uses the lookup table read from the memory unit 3 to correct the apparent radiance of the object A, thereby calculating the true radiance for each pixel.

[0114] Next, the emissivity correction unit 27 estimates the infrared emissivity of the object A, and multiplies the luminance image by the estimated emissivity to generate a temperature image, which is a luminance image that has been subjected to luminance correction using the emissivity (step ST4A-5).

[0115] The infrared filter selection unit 28 uses the temperature image output from the emissivity correction unit 27 to determine whether or not the current combination of the infrared filter unit 5A and the infrared filter unit in the infrared filter changing unit 5-1 is optimal for the current temperature of the object A to be measured (step ST4A-6). If it is determined that the combination of the infrared filter units is optimal (step ST4A-6; YES), the series of processes in FIG. 9 ends.

[0116] When it is determined that the combination of infrared filter parts is not optimal (step ST4A-6; NO), the infrared filter selection unit 28 changes the infrared filter part set in the infrared filter changing unit 5-1 (step ST4A-7).

[0117] It is known that the shape of the blackbody radiation spectrum changes depending on the temperature of the object. In the emissivity correction in step ST4A-5, the shape of the spectrum is estimated by capturing the radiation from the object A using infrared camera 4A and infrared camera 4B, which have different light-receiving sensitivity wavelength bands, and the emissivity and temperature are estimated using the radiation spectrum. The processes in steps ST4A-6 and ST4A-7 are intended to improve the accuracy of temperature measurement by selecting an optimum combination of infrared light sensitivity wavelength bands in accordance with the temperature of object A to be measured.

[0118] In the emissivity correction, the radiance of a luminance image based on the infrared images captured by the infrared cameras 4A and 4B is used to determine the temperature and emissivity of the object A. Therefore, if the radiance in the luminance image can be determined accurately, the accuracy of estimating the temperature and emissivity of the object A will increase accordingly.

[0119] Infrared cameras have random noise, such as NETD, and a certain amount of error is superimposed on the measured radiance. Considering this, it is necessary to select the wavelength bands at which the radiance of the luminance image obtained using infrared camera 4A and the radiance of the luminance image obtained using infrared camera 4B are maximized.

[0120] For example, there is a method of setting the sensitivity wavelength bands of the infrared cameras 4A and 4B so that they are symmetrical on either side of the peak wavelength. This allows the radiance of the luminance images obtained using the infrared cameras 4A and 4B to be roughly maximized. According to Wien's displacement meter, the peak wavelength λ of the blackbody radiation spectrum at temperature T is p is calculated using the following formula (4). λ p ≒(2.898×10 -3 ) / T (4)

[0121] FIG. 10 is a graph showing the characteristics of the peak wavelength of the blackbody radiation spectrum, and shows the peak wavelength calculated by the above formula (4). According to the above formula (4), as shown in FIG. 10, when the temperature of the object A to be measured is 305K, the peak wavelength of the blackbody radiation spectrum is about 9.5 μm. Therefore, the optimal combination of center wavelengths of the infrared filter unit is 9 μm-10 μm. Also, when the temperature of the object A to be measured is 340K, the peak wavelength is about 8.5 μm, so the optimal combination of center wavelengths of the infrared filter unit is 8 μm-9 μm.

[0122] The above-described method of setting the center wavelength of the infrared filter is just one example, and the characteristics of the infrared filter section and their combination that maximize the accuracy of temperature measurement will change depending on the assumed transmission bandwidth of the infrared filter section or temperature measurement conditions. For example, a lookup table showing an optimum filter combination for the temperature of the object A to be measured may be prepared, and the lookup table may be used to determine whether the combination of infrared filter units in step ST4A-6 is optimum.

[0123] The functions of the control unit 21, the sensitivity wavelength control unit 22, the luminance calibration unit 23, the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, the emissivity correction unit 27, and the infrared filter selection unit 28 included in the non-contact temperature measurement device 1A may be realized by a processing circuit 102 of dedicated hardware shown in Fig. 6A. The processing circuit 102 corresponds to, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination of these. The functions of the control unit 21, the sensitivity wavelength control unit 22, the luminance calibration unit 23, the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, the emissivity correction unit 27, and the infrared filter selection unit 28 included in the non-contact temperature measurement device 1A may be realized by separate processing circuits, or these functions may be realized together by one processing circuit.

[0124] 6B, the functions of the control unit 21, sensitivity wavelength control unit 22, luminance calibration unit 23, matching processing unit 24, positional relationship estimation unit 25, luminance correction unit 26, emissivity correction unit 27 and infrared filter selection unit 28 included in the non-contact temperature measurement device 1A are realized by software, firmware or a combination of software and firmware. The software or firmware is written as a program and stored in the memory 104.

[0125] The processor 103 reads out and executes the programs stored in the memory 104, thereby realizing the functions of the control unit 21, the sensitivity wavelength control unit 22, the luminance calibration unit 23, the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, the emissivity correction unit 27 and the infrared filter selection unit 28 provided in the non-contact temperature measuring device 1A. For example, the non-contact temperature measuring device 1A includes a memory 104 for storing a program that, when executed by the processor 103, results in the processing of steps ST4A-1 to ST4A-7 shown in FIG. These programs cause the computer to execute the procedures or methods of processing performed by the control unit 21, the sensitivity wavelength control unit 22, the luminance calibration unit 23, the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, the emissivity correction unit 27, and the infrared filter selection unit 28. The memory 104 may be a computer-readable storage medium having stored therein programs for causing the computer to function as the control unit 21, the sensitivity wavelength control unit 22, the luminance calibration unit 23, the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, the emissivity correction unit 27, and the infrared filter selection unit 28.

[0126] As described above, the non-contact temperature measuring device 1A according to the second embodiment includes the infrared filter selection unit 28. The infrared filter selection unit 28 selects a combination of light sensitivity wavelength bands to be used for measuring the temperature of the object A by using a luminance image obtained by the luminance correction unit 26 and the emissivity correction unit 27 correcting the luminance value of each pixel in an infrared image captured by the infrared camera 4A and the infrared camera 4B using a combination of two or more light sensitivity wavelength bands to capture a field of view including the same scene. The non-contact temperature measuring device 1A can improve the accuracy of temperature measurement by selecting an optimal combination of infrared light sensitivity wavelength bands according to the temperature of the object A to be measured.

[0127] Embodiment 3 Fig. 11 is a block diagram showing a configuration example of a non-contact temperature measuring device 1B according to embodiment 3. In Fig. 11, the non-contact temperature measuring device 1B is a device that measures the temperature of an object A in a non-contact manner by using an infrared image of the object A whose temperature is to be measured, and includes a signal processing unit 2B, a memory unit 3, an infrared camera 4A, an infrared camera 4B, an infrared filter unit 5A, an infrared filter changing unit 5-1, and a shutter unit 6.

[0128] The signal processing unit 2B performs temperature measurement processing of the object A using the infrared images captured by the infrared cameras 4A and 4B. The signal processing unit 2B includes a control unit 21, a sensitivity wavelength control unit 22, a luminance calibration unit 23, a matching processing unit 24, a positional relationship estimation unit 25, a luminance correction unit 26, an emissivity correction unit 27, an infrared filter selection unit 28, and a temperature image synthesis unit 29.

[0129] The temperature image synthesis unit 29 is an image synthesis unit that synthesizes a plurality of luminance images, which are luminance images based on infrared images captured by the infrared cameras 4A and 4B of a field of view including a common area using the combination of light sensitivity wavelength bands selected by the infrared filter selection unit 28, and in which the radiance of each pixel has been corrected by the luminance correction unit 26 and the emissivity correction unit 27. For example, the temperature image synthesis unit 29 acquires a temperature image obtained using the current combination of infrared filter sections from the infrared filter selection unit 28, reads out a temperature image obtained using a combination of infrared filter sections different from the current one from the memory unit 3, and generates an image by synthesizing these temperature images as a temperature image with improved accuracy of the measured temperature.

[0130] Although the signal processing unit 2B has been shown to include the control unit 21, the sensitivity wavelength control unit 22, the luminance calibration unit 23, the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, the emissivity correction unit 27, the infrared filter selection unit 28, and the temperature image synthesis unit 29, it is sufficient for the non-contact temperature measurement device 1B to include at least the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, the emissivity correction unit 27, the infrared filter selection unit 28, and the temperature image synthesis unit 29. In this case, the control unit 21, the sensitivity wavelength control unit 22, and the luminance calibration unit 23 are provided in, for example, the infrared camera 4A or the infrared camera 4B.

[0131] Next, a specific operation will be described. Fig. 12 is a flow chart showing the operation of the non-contact temperature measuring device 1B, and shows a series of processes for temperature measurement by the non-contact temperature measuring device 1B. In Fig. 12, the processes of steps ST1, ST2, ST3, and ST5 are similar to the processes of steps ST1, ST2, ST3, and ST5 in Fig. 2, and therefore description thereof will be omitted.

[0132] The luminance correction unit 26, the emissivity correction unit 27, and the temperature image synthesis unit 29 perform temperature correction and temperature image synthesis on the luminance image for which the stereo correction has been completed (step ST4B). Temperature image synthesis is a process of synthesizing temperature images acquired by combining multiple infrared filter units. Temperature correction is a process of correcting the influence of the distance between object A and infrared cameras 4A and 4B and the infrared emissivity of object A on the brightness image based on the infrared images captured by infrared cameras 4A and 4B.

[0133] FIG. 13 is a flowchart showing a series of processes for temperature correction and temperature image synthesis, and shows a series of processes in step ST4B of FIG. The control unit 21 outputs sensitivity wavelength band information to the infrared filter changing unit 5-1, which specifies the transmission wavelength band of the infrared filter unit 5C or the infrared filter unit 5D as the sensitivity wavelength so that the infrared cameras 4A and 4B have different light receiving sensitivity wavelength bands. The sensitivity wavelength control unit 22 outputs a control signal for switching the infrared filter unit to the infrared filter changing unit 5 based on the sensitivity wavelength band information acquired from the control unit 21. The infrared filter changing unit 5 changes the infrared filter unit to be set in the field of view B2 of the infrared camera 4B to the infrared filter unit 5C or the infrared filter unit 5D in accordance with the control signal (step ST4B-1).

[0134] With the infrared filter unit 5A set in the infrared camera 4A and the infrared filter unit 5C or 5D set in the infrared camera 4B, the infrared camera 4A captures an image of the field of view B1, and the infrared camera 4B captures an image of the field of view B2. The control unit 21 acquires the infrared images captured by the infrared cameras 4A and 4B (step ST4B-2). The infrared image acquired by the control unit 21 is a multiband image captured by the infrared cameras 4A and 4B in different light receiving sensitivity wavelength bands.

[0135] The control unit 21 outputs the infrared images acquired from the infrared cameras 4A and 4B and instruction information for performing luminance calibration to convert these infrared images into luminance images to the luminance calibration unit 23. The luminance calibration unit 23 reads a calibration table from the memory unit 3 according to the instruction information, and converts the pixel values ​​of the infrared images into radiance using luminance calibration parameters corresponding to different light receiving sensitivity wavelength bands included in the calibration table, thereby generating a luminance image (step ST4B-3). The luminance calibration unit 23 outputs the generated luminance image to the luminance correction unit 26.

[0136] The luminance correction unit 26 uses the distance map read from the memory unit 3 to correct the radiance of each pixel of the luminance image acquired from the luminance calibration unit 23 (step ST4B-4). For example, the luminance correction unit 26 uses the lookup table read from the memory unit 3 to correct the apparent radiance of the object A, thereby calculating the true radiance for each pixel.

[0137] Next, the emissivity correction unit 27 estimates the infrared emissivity of the object A, and multiplies the luminance image by the estimated emissivity to generate a temperature image, which is a luminance image that has been subjected to luminance correction using the emissivity (step ST4B-5).

[0138] The temperature image synthesis unit 29 judges whether or not the temperature of the object A has been measured with a combination of multiple infrared filter units (step ST4B-6). If it is judged that the temperature of the object A has not been measured with a combination of multiple infrared filter units (step ST4B-6; NO), the infrared filter selection unit 28 changes the infrared filter unit set in the infrared filter change unit 5-1 (step ST4B-7). With the changed combination of infrared filter units, the series of processes from step ST4B-2 are repeated.

[0139] If it is determined that the temperature of object A has been measured using multiple combinations of infrared filter sections (step ST4B-6; YES), the temperature image synthesis unit 29 synthesizes the multiple temperature images acquired using the multiple combinations of infrared filter sections to generate a temperature image having high measurement accuracy across the entire image (step ST4B-8).

[0140] In the second embodiment, the combination of infrared filters is changed according to the temperature of the object A to be measured, and the measurement is performed in the optimum sensitivity wavelength band for that temperature, improving the accuracy of the temperature measurement. However, with this method, if there are objects with significantly different temperatures in the image, the combination of infrared filters can only be optimized for one of the objects.

[0141] In contrast, in the temperature image synthesis in step ST4B-8, multiple temperature images acquired by combining multiple infrared filter units are synthesized. For example, a certain temperature threshold is set, and for pixels below the threshold, the temperature value acquired by infrared filter unit combination (1) is used, and for pixels above the threshold, the temperature value acquired by infrared filter unit combination (2) is used, and these are synthesized into one image. This allows the non-contact temperature measuring device 1B to set an optimal combination of infrared filter units for each pixel, improving the accuracy of temperature measurement for the entire image. Note that two or more of the above thresholds may be set.

[0142] The functions of the control unit 21, the sensitivity wavelength control unit 22, the luminance calibration unit 23, the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, the emissivity correction unit 27, the infrared filter selection unit 28 and the temperature image synthesis unit 29 provided in the non-contact temperature measuring device 1B may be realized by a processing circuit 102 of dedicated hardware shown in FIG. 6A. The processing circuit 102 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination of these. The functions of the control unit 21, the sensitivity wavelength control unit 22, the luminance calibration unit 23, the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, the emissivity correction unit 27, the infrared filter selection unit 28, and the temperature image synthesis unit 29 included in the non-contact temperature measurement device 1B may be realized by separate processing circuits, or these functions may be realized together by one processing circuit.

[0143] 6B, the functions of the control unit 21, sensitivity wavelength control unit 22, luminance calibration unit 23, matching processing unit 24, positional relationship estimation unit 25, luminance correction unit 26, emissivity correction unit 27, infrared filter selection unit 28 and temperature image synthesis unit 29 included in the non-contact temperature measurement device 1B are realized by software, firmware or a combination of software and firmware. The software or firmware is written as a program and stored in the memory 104.

[0144] The processor 103 reads out and executes the programs stored in the memory 104, thereby realizing the functions of the control unit 21, the sensitivity wavelength control unit 22, the luminance calibration unit 23, the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, the emissivity correction unit 27, the infrared filter selection unit 28 and the temperature image synthesis unit 29 provided in the non-contact temperature measuring device 1B. For example, the non-contact temperature measuring device 1B includes a memory 104 for storing a program that, when executed by the processor 103, results in the processing of steps ST4B-1 to ST4B-8 shown in FIG. These programs cause the computer to execute the procedures or methods of processing performed by the control unit 21, the sensitivity wavelength control unit 22, the luminance calibration unit 23, the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, the emissivity correction unit 27, the infrared filter selection unit 28 and the temperature image synthesis unit 29. The memory 104 may be a computer-readable storage medium having stored therein programs for causing a computer to function as the control unit 21, the sensitivity wavelength control unit 22, the luminance calibration unit 23, the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, the emissivity correction unit 27, the infrared filter selection unit 28, and the temperature image synthesis unit 29.

[0145] As described above, the non-contact temperature measuring device 1B according to the third embodiment includes the temperature image synthesis unit 29. The temperature image synthesis unit 29 synthesizes a plurality of images obtained by correcting the luminance of each pixel in a plurality of infrared images captured by the infrared camera 4A and the infrared camera 4B of a field of view including the same scene using the combination of light receiving sensitivity wavelength bands selected by the infrared filter selection unit 28, using the luminance correction unit 26 and the emissivity correction unit 27. This allows the non-contact temperature measuring device 1B to set an optimal combination of infrared filter sections for each pixel, thereby improving the temperature measurement accuracy over the entire image.

[0146] Embodiment 4 Fig. 14 is a block diagram showing a configuration example of a non-contact temperature measuring device 1C according to embodiment 4. In Fig. 14, the non-contact temperature measuring device 1C is a device that measures the temperature of an object A in a non-contact manner by using an infrared image of the object A whose temperature is to be measured, and includes a signal processing unit 2C, a memory unit 3, an infrared camera 4A, an infrared camera 4B, an infrared filter unit 5A, an infrared filter changing unit 5-1, a shutter unit 6, a reference temperature measuring unit 7, a housing 8, an infrared camera 9 for shutter luminance measurement, and a reference target 10.

[0147] The signal processing unit 2C performs temperature measurement processing of the object A using the infrared images captured by the infrared cameras 4A and 4B. The signal processing unit 2C includes a control unit 21, a sensitivity wavelength control unit 22A, a luminance calibration unit 23, a matching processing unit 24, a positional relationship estimation unit 25, a luminance correction unit 26, an emissivity correction unit 27, an infrared filter selection unit 28, and a temperature image synthesis unit 29.

[0148] The reference temperature measurement unit 7 measures the temperature of a reference target 10 provided separately from the object A whose temperature is to be measured. For example, the reference temperature measurement unit 7 is a contact-type temperature sensor provided near the reference target 10, measures the temperature of the reference target 10, and outputs the measured temperature to the brightness calibration unit 23. As the temperature sensor, for example, a thermistor or a thermocouple may be used.

[0149] The housing 8 accommodates therein the signal processing unit 2C, the memory unit 3, the infrared camera 4A, the infrared camera 4B, the infrared filter unit 5A, the infrared filter changing unit 5-1, the shutter unit 6, the reference temperature measurement unit 7, the housing 8, the infrared camera 9 for shutter luminance measurement, and the reference target 10. The reference target 10 is held on the surface of the housing 8. The housing 8 may be divided. For example, the signal processing unit 2C, the infrared camera 4A, and the infrared camera 4B may each have a separate housing.

[0150] The infrared camera 9 for shutter luminance measurement captures an infrared image of the shutter of the shutter unit 6 arranged in front of the infrared cameras 4A and 4B, and captures an infrared image of a reference target 10 provided on the surface of the housing 8. The infrared camera 9 for shutter luminance measurement outputs the captured infrared image to the luminance calibration unit 23.

[0151] Further, the infrared camera 9 for measuring shutter luminance is composed of, for example, a single infrared camera separate from the infrared cameras 4A and 4B. The infrared camera 9 for measuring shutter luminance is installed outside the range of the field of view B1 and the field of view B2 so as not to be photographed by the infrared cameras 4A and 4B. The infrared camera 9 for measuring shutter luminance has a field of view at least twice as large as that of the infrared cameras 4A and 4B so as to include the two shutters of the shutter unit 6 and the reference target 10 within its field of view from its installation position. The shutter luminance measuring infrared camera 9 may be provided independently of the housing 8, or may be supported by the housing 8 using a plate-like structure or the like.

[0152] Although the signal processing unit 2C has been shown to have a configuration including the control unit 21, the sensitivity wavelength control unit 22, the luminance calibration unit 23, the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, the emissivity correction unit 27, the infrared filter selection unit 28, and the temperature image synthesis unit 29, it is sufficient for the non-contact temperature measurement device 1C to include at least the matching processing unit 24, the positional relationship estimation unit 25, the luminance correction unit 26, the emissivity correction unit 27, the infrared filter selection unit 28, and the temperature image synthesis unit 29. In this case, the control unit 21, the sensitivity wavelength control unit 22, and the luminance calibration unit 23 are provided in, for example, the infrared camera 4A or the infrared camera 4B.

[0153] Next, a specific operation will be described. Fig. 15 is a flow chart showing the operation of the non-contact temperature measuring device 1C, and shows a series of processes for temperature measurement by the non-contact temperature measuring device 1C. In Fig. 15, the processes of steps ST1, ST3, ST4, and ST5 are similar to the processes of steps ST1, ST3, ST4, and ST5 in Fig. 2, and therefore description thereof will be omitted.

[0154] When a certain time has passed since the previous shutter correction, the signal processing unit 2C performs a high-precision shutter correction (step ST2A). The high-precision shutter correction measures the temperature of the shutter and the temperature of the reference target 10 using a luminance image of the shutter of the shutter unit 6 and a luminance image of the reference target 10, and estimates with high precision the radiance of the luminance image according to the temperature of the shutter using the temperature of the reference target 10 measured by the reference temperature measuring unit 7 as a known value.

[0155] FIG. 16 is a flowchart showing a series of processes for high-precision shutter correction, and shows a series of processes in step ST2A of FIG. First, the control unit 21 transmits a control signal for closing the shutters to the shutter unit 6. The shutter unit 6 closes the shutters in accordance with the control signal, blocking the fields of view of the infrared cameras 4A and 4B (step ST2A-1).

[0156] Next, the infrared cameras 4A and 4B capture images of the shutters blocking the visual fields B1 and B2 to generate shutter images. The control unit 21 acquires the shutter images from the infrared cameras 4A and 4B (step ST2A-2).

[0157] The shutter luminance measurement in step ST2A-3 is a process in which the luminance calibration unit 23 converts the infrared images of the two shutters provided in the shutter unit 6 captured by the shutter luminance measurement infrared camera 9 into luminance images. Next, the reference plate brightness and temperature measurement in step ST2A-4 is a process in which the brightness calibration unit 23 converts the infrared image of the reference target 10 captured by the shutter brightness measurement infrared camera 9 into a brightness image, and further, the reference temperature measurement unit 7 measures the temperature of the reference target 10.

[0158] The luminance calibration unit 23 converts the radiance in the shutter image into a temperature using the luminance image of the shutter, the luminance image of the reference target 10, and the measured temperature of the reference target 10 (step ST2A-5).

[0159] Next, the brightness calibration unit 23 obtains infrared images of object A captured by the infrared cameras 4A and 4B from the control unit 21, and reads out from the memory unit 3 sensitivity wavelength band information indicating the light receiving sensitivity wavelength bands of the infrared cameras 4A and 4B, transmission wavelength band information indicating the transmission wavelength bands of the infrared filter units currently set therefor, and the like. Then, using the information acquired from the control unit 21 and the information read from the memory unit 3, the brightness calibration unit 23 calculates brightness calibration parameters required for converting the pixel values ​​of the infrared image into the radiance of object A for each pixel of the infrared image (step ST2A-6). The luminance calibration unit 23 generates a calibration table in which parameters for luminance calibration for each pixel are set, and stores the generated calibration table in the memory unit 3.

[0160] Next, the control unit 21 outputs the sensitivity wavelength band information to the sensitivity wavelength control unit 22 to specify the light receiving sensitivity wavelength band. The sensitivity wavelength control unit 22 outputs a control signal to the infrared filter changing unit 5 to switch to an infrared filter unit that transmits the wavelength band indicated by the sensitivity wavelength band information acquired from the control unit 21. The infrared filter changing unit 5-1 changes the infrared filter unit set in the field of view B2 of the infrared camera 4B to the infrared filter unit 5B or the infrared filter unit 5C in accordance with the control signal (step ST2A-7).

[0161] After changing the infrared filter unit, the control unit 21 returns to the process of step ST2A-2 and performs the processes of steps ST2A-2 and ST2A-6 again. This series of processes is performed for all the infrared filter units included in the infrared filter changing unit 5-1.

[0162] When the above-mentioned processing is completed, the control unit 21 transmits a control signal for opening the shutter to the shutter unit 6. The shutter unit 6 opens the shutter in accordance with the control signal, and ends blocking the fields of view of the infrared cameras 4A and 4B (step ST2A-8).

[0163] The shutter correction in step ST2A-1 is a common method used in infrared cameras. Even if the temperature of the object A to be measured is constant, the apparent temperature (brightness) fluctuates due to changes in the temperature of the infrared camera itself. For this reason, the brightness of the shutter, whose temperature can be considered known, is measured to associate the measured pixel value with the radiance value of the object to be measured. At this time, the shutter temperature must be known, but the shutter temperature is usually estimated using the measurement value of a contact-type temperature sensor attached to the shutter, or by approximating it to be approximately equal to room temperature.

[0164] On the other hand, since the shutter used for correction is a part that opens and closes, it is not possible to contact the temperature sensor with the shutter itself, and the temperature that can be measured is limited to the temperature around the shutter. For this reason, the measurement value with a contact-type temperature sensor is not the temperature of the shutter itself. The shutter temperature is affected by a complex range of environmental temperatures, including heat from the infrared camera itself, so it does not necessarily match the room temperature. Also, it is not possible to take into account the temperature distribution of the shutter.

[0165] In contrast, the non-contact temperature measuring device 1C estimates temperature and emissivity by combining brightness values ​​measured by multiple infrared cameras. At this time, various calculations are added to the temperature measurement values ​​using each infrared camera, so if there is an error in these measurement values, the error in the measured temperature and emissivity output also increases. Therefore, the infrared camera is used as a means to measure the shutter temperature with higher accuracy than when used for general purposes.

[0166] Note that a single infrared camera may be used as the shutter luminance measurement infrared camera 9. In this case, errors due to individual differences in the infrared camera can be eliminated, making it possible to measure the shutter temperature with higher accuracy.

[0167] Moreover, a high-resolution infrared camera may be used as the shutter luminance measurement infrared camera 9. In this case, it is possible to acquire a two-dimensional temperature distribution of the shutter, and it becomes possible to calibrate the infrared cameras 4A and 4B with higher accuracy.

[0168] As described above, the non-contact temperature measuring device 1C according to the fourth embodiment includes the reference temperature measuring unit 7 that measures the temperature of the reference target 10 provided separately from the object A whose temperature is to be measured. The luminance calibration unit 23 uses an infrared image including the reference target 10 and the shutter unit 6 captured by the shutter luminance measurement infrared camera 9 to generate a luminance image showing a distribution of luminance values ​​for each pixel corresponding to the temperatures of the reference target 10 and the shutter unit 6. The luminance correction unit 26 corrects the luminance image of the shutter unit 6 using the correspondence between the luminance image of the reference target 10 and the temperature measurement value of the reference target 10, and corrects the luminance image of the object A using the corrected luminance image of the shutter unit 6. This improves the accuracy of the luminance calibration in each infrared camera in the non-contact temperature measurement device 1C, and as a result, improves the accuracy of the measured temperature.

[0169] It is possible to combine the embodiments, modify any of the components of the embodiments, or omit any of the components of the embodiments. [Industrial Applicability]

[0170] The non-contact temperature measuring device according to the present disclosure can be used, for example, for remote monitoring of a person with a fever. [Explanation of symbols]

[0171] 1, 1A to 1C non-contact temperature measuring device, 2, 2A to 2C signal processing unit, 3 memory unit, 4A, 4B infrared camera, 5, 5-1, 5A to 5D infrared filter changing unit, 6 shutter unit, 7 reference temperature measurement unit, 8 housing, 9 infrared camera for shutter brightness measurement, 10 reference target, 21 control unit, 22, 22A sensitivity wavelength control unit, 23 brightness calibration unit, 24 matching processing unit, 25 positional relationship estimation unit, 26 brightness correction unit, 27 emissivity correction unit, 28 infrared filter selection unit, 29 temperature image synthesis unit, 100 input interface, 101 output interface, 102 processing circuit, 103 processor, 104 memory.

Claims

1. a matching processing unit that estimates pixel correspondences between a plurality of images captured by a plurality of infrared cameras of a field of view including a common area; a wavelength sensitivity band changing unit that changes the wavelength sensitivity band of at least one of the infrared cameras; a positional relationship estimation unit that estimates a positional relationship between the infrared camera and an object that is a temperature measurement target and exists in the common area based on the estimated pixel correspondence relationship; a pixel value correction unit that corrects pixel values ​​corresponding to temperature based on the estimated positional relationship between the infrared camera and the object; an emissivity correction unit that estimates an infrared emissivity of the object by using the estimated pixel correspondence and the image in which the pixel values ​​have been corrected, and generates an image in which the pixel values ​​have been corrected based on the estimated emissivity; the sensitivity wavelength band changing unit changes the light sensitivity wavelength band of at least one of the infrared cameras so that the light sensitivity wavelength bands of the infrared cameras are the same or different from each other; the matching processing unit estimates a correspondence relationship between pixels between images captured by the plurality of infrared cameras of a field of view including the common area when the light receiving sensitivity wavelength bands of at least two of the plurality of infrared cameras are the same; The emissivity correction unit estimates the emissivity of infrared light of the object by using images captured by the plurality of infrared cameras of a field of view including the common area when at least two of the plurality of infrared cameras have different light receiving sensitivity wavelength bands. A non-contact temperature measuring device characterized by the above.

2. The sensitivity wavelength control section an infrared filter change unit that has a plurality of infrared filter units through which infrared light of different wavelength bands pass, and switches the infrared filter unit that passes infrared light incident on at least one of the plurality of infrared cameras; a sensitivity wavelength control unit that controls switching of the infrared filter unit, The sensitivity wavelength control unit controls the switching of the infrared filter unit by the infrared filter changing unit to match the light receiving sensitivity wavelength bands of at least two of the infrared cameras.

2. The non-contact temperature measuring device according to claim 1.

3. A shutter unit that blocks the field of view of the multiple infrared cameras; a brightness calibration unit that converts a plurality of infrared images captured by the plurality of infrared cameras into a brightness image that indicates a distribution of brightness values ​​for each pixel corresponding to temperature.

3. The non-contact temperature measuring device according to claim 2.

4. The light receiving sensitivity wavelength bands of the plurality of infrared cameras are a wavelength band of 8 to 14 micrometers, a wavelength band of 3 to 5 micrometers, and a wavelength band that is different from these wavelength bands by a certain wavelength.

4. The non-contact temperature measuring device according to claim 3.

5. The sensitivity wavelength control unit controls the switching of the infrared filter unit by the infrared filter changing unit to match the light receiving sensitivity wavelength bands of at least two of the infrared cameras.

5. The non-contact temperature measuring device according to claim 4.

6. The shutter unit is configured integrally with the infrared filter changing unit, and blocks the field of view of the infrared camera into which infrared light that has passed through the infrared filter unit switched by the infrared filter changing unit is incident.

6. The non-contact temperature measuring device according to claim 5.

7. The sensitivity wavelength control unit is capable of changing the number of combinations of the light receiving sensitivity wavelength bands of the plurality of infrared cameras to at least three or more.

6. The non-contact temperature measuring device according to claim 5.

8. At least two of the combinations of the light sensitivity wavelengths of the infrared cameras are different from each other in the combinations of the light sensitivity wavelengths of the infrared cameras, and an infrared filter selection unit that selects a combination of light receiving sensitivity wavelength bands to be used for measuring the temperature of the object by using an image obtained by correcting the pixel values ​​in an infrared image captured by the plurality of infrared cameras of a field of view including the common region by the pixel value correction unit and the emissivity correction unit.

8. The non-contact temperature measuring device according to claim 7.

9. and an image synthesis unit that synthesizes a plurality of images obtained by correcting pixel values ​​in a plurality of infrared images captured by a plurality of the infrared cameras of a field of view including the common region using a combination of light receiving sensitivity wavelength bands selected by the infrared filter selection unit, the pixel value correction unit and the emissivity correction unit. The non-contact temperature measuring device according to claim 8 .

10. a reference temperature measurement unit that measures the temperature of a reference target provided separately from the object to be temperature-measured; The luminance calibration unit generating a luminance image showing a distribution of luminance values ​​for each pixel corresponding to temperatures of the reference target and the shutter section using an infrared image including the reference target and the shutter section captured by an infrared camera for shutter luminance measurement provided separately from the plurality of infrared cameras; The pixel value correction unit correcting the luminance image of the shutter unit using a correspondence relationship between the luminance image of the reference target and the temperature measurement value of the reference target; The luminance image of the object is corrected using the corrected luminance image of the shutter section.

8. The non-contact temperature measuring device according to claim 7.

11. A non-contact temperature measurement method using a non-contact temperature measurement device, comprising: A matching processing unit estimates pixel correspondences between a plurality of images captured by a plurality of infrared cameras of a field of view including a common region; A sensitivity wavelength band changing unit changes the sensitivity wavelength band of at least one of the infrared cameras; a positional relationship estimation unit estimating a positional relationship between the infrared camera and an object that is a temperature measurement target and exists in the common area based on the estimated pixel correspondence; a pixel value correction unit correcting a pixel value corresponding to a temperature based on the estimated positional relationship between the infrared camera and the object; an emissivity correction unit estimating an infrared emissivity of the object using the estimated pixel correspondence and the image in which the pixel values ​​have been corrected, and generating an image in which the pixel values ​​have been corrected based on the estimated emissivity; the sensitivity wavelength band changing unit changes the light sensitivity wavelength band of at least one of the infrared cameras so that the light sensitivity wavelength bands of the infrared cameras are the same or different from each other; the matching processing unit estimates a correspondence relationship between pixels between images captured by the plurality of infrared cameras of a field of view including the common area when the light receiving sensitivity wavelength bands of at least two of the plurality of infrared cameras are the same; The emissivity correction unit estimates the emissivity of infrared light of the object by using images captured by the plurality of infrared cameras of a field of view including the common area when at least two of the plurality of infrared cameras have different light receiving sensitivity wavelength bands. A non-contact temperature measurement method comprising: