JP7371645B2 - Surface temperature measurement system and surface temperature measurement method - Google Patents

Description

The present invention relates to a technique for measuring surface temperature without contact.

Due to the impact of COVID-19, demand for non-contact body temperature measurement is increasing. Conventionally, thermography using a bolometer method has been used at airports, etc., but this system is extremely expensive, so it is difficult to introduce it into commercial facilities, offices, etc. Therefore, in applications such as commercial facilities and offices, surface temperature measurement systems using thermopile type thermal sensors are widely used (see Patent Document 1).

Japanese Patent Application Publication No. 2013-200137

Thermopile type thermal sensors have a problem in that the output temperature changes depending on the relative position of the sensor and the object to be measured (such as a face) due to the influence of the pixel size and the influence of the lens. As a measure to solve such problems, Patent Document 1 proposes the idea of measuring the distance to the object to be measured using a TOF sensor and correcting the output temperature of the thermal sensor based on the distance. However, the following problems remain with the measure of Patent Document 1, as described below.

The inventors conducted experiments and found that the output temperature of a thermal sensor depends not only on the distance between the sensor and the object to be measured in the optical axis direction, but also on the amount of positional deviation of the object to be measured with respect to the optical axis. It turns out that things are changing. That is, when the object to be measured is on the optical axis, accurate measured values can be obtained, but when the object to be measured moves away from the optical axis, the measured values fluctuate. The method disclosed in Patent Document 1 does not take into account the positional shift of the object to be measured with respect to the optical axis, so there is a limit to the correction accuracy, and there are cases where accurate surface temperature cannot be obtained.

The present invention has been made in view of the above circumstances, and an object thereof is to provide a technique that can more accurately measure the surface temperature of an object using a thermopile type thermal sensor.

The present disclosure provides a thermopile type thermal sensor that measures the surface temperature of a measurement target, a distance between the thermal sensor and the measurement target in the optical axis direction of the thermal sensor, and a distance between the thermal sensor and the measurement target in the optical axis direction of the thermal sensor. The surface temperature measurement system includes a processing unit that corrects the measured value of the thermal sensor based on the amount of positional shift of the measurement target. By adopting such a configuration, it is possible to realize a system that can accurately measure the surface temperature of the object to be measured while using a thermopile type sensor.

The apparatus may further include a camera having a common field of view with the thermal sensor, and the processing unit may calculate the amount of positional deviation of the measurement object with respect to the optical axis by analyzing an image taken by the camera. The processing unit may detect the object to be measured from the image through object detection processing, and obtain a positional shift amount of the object to be measured with respect to the optical axis based on a detected position on the image. The processing unit may determine the distance to the object to be measured by analyzing an image taken by the camera. The processing unit may detect the object to be measured from the image by object detection processing, and calculate the distance of the object to be measured based on the detected size on the image.

The processing unit further includes a camera having a common field of view with the thermal sensor, and when the image taken by the camera and the measurement value output from the thermal sensor are input, the processing unit calculates the distance and positional deviation of the object to be measured. The measured value of the thermal sensor may be corrected using AI that has been machine learned to perform correction according to the amount.

The processing unit may determine the amount of positional deviation of the measurement object with respect to the optical axis by analyzing a temperature distribution image output from the thermal sensor. The processing unit may detect the object to be measured from the temperature distribution image through object detection processing, and obtain a positional shift amount of the object to be measured with respect to the optical axis based on a detected position on the temperature distribution image. . The processing unit may determine the distance to the measurement target by analyzing the temperature distribution image. The processing unit may detect the object to be measured from the temperature distribution image by object detection processing, and calculate the distance of the object to be measured based on the detected size on the temperature distribution image.

The processing unit processes the thermal sensor using AI that has been machine-learned to perform correction according to the distance and positional deviation amount of the measurement target when the temperature distribution image output from the thermal sensor is input. The measured value of may be corrected.

The processing unit further includes a TOF sensor having a common field of view with the thermal sensor, and the processing unit determines the distance of the measurement target and the measurement target with respect to the optical axis by analyzing a distance image obtained by the TOF sensor. The amount of positional shift of the object may also be determined.

The present disclosure includes a step of measuring a surface temperature of an object to be measured using a thermopile type thermal sensor, a distance between the thermal sensor and the object to be measured in the optical axis direction of the thermal sensor, and a distance between the thermal sensor and the object to be measured in the optical axis direction of the thermal sensor. The method includes the step of correcting the measured value of the thermal sensor based on the amount of positional deviation of the object to be measured with respect to the optical axis.

The present invention may be regarded as a surface temperature measurement system having at least a part of the above means, or a surface temperature measurement method including at least a part of the above processing, or a program for realizing such a method, or a program thereof. It can also be regarded as a recorded recording medium. Note that each of the above means and processes can be combined to the extent possible to constitute the present invention.

According to the present invention, it is possible to provide a technique that can more accurately measure the surface temperature of an object using a thermopile type thermal sensor.

FIG. 1 is a diagram showing an application example of a surface temperature measurement system. FIG. 2 is a perspective view showing the appearance of the surface temperature measurement system of the first embodiment. FIG. 3 is a block diagram showing the hardware configuration of the surface temperature measurement system of the first embodiment. FIG. 4 is a cross-sectional view schematically showing the configuration of the thermal sensor. FIG. 5 is a block diagram showing the functional configuration of the surface temperature measurement system of the first embodiment. FIG. 6 is a flowchart showing the flow of surface temperature measurement processing by the surface temperature measurement system of the first embodiment. FIG. 7 is a diagram showing an example of face detection. FIG. 8 is a diagram illustrating the positional relationship between the optical axis and field of view of the sensor and the measurement target. FIGS. 9A and 9B are diagrams showing examples of output of measurement results from the surface temperature measurement system. FIG. 10 is a block diagram showing the functional configuration of the surface temperature measurement system of the second embodiment. FIG. 11 is a flowchart showing the flow of surface temperature measurement processing by the surface temperature measurement system of the second embodiment. FIG. 12 is a block diagram showing the hardware configuration of the surface temperature measurement system of the third embodiment. FIG. 13 is a block diagram showing the functional configuration of the surface temperature measurement system of the third embodiment. FIG. 14 is a flowchart showing the flow of surface temperature measurement processing by the surface temperature measurement system of the third embodiment. FIG. 15 is a block diagram showing the hardware configuration of the surface temperature measurement system of the fourth embodiment. FIG. 16 is a block diagram showing the functional configuration of the surface temperature measurement system of the fourth embodiment. FIG. 17 is a flowchart showing the flow of surface temperature measurement processing by the surface temperature measurement system of the fourth embodiment.

<Application example>
One application example of the present invention will be described with reference to FIG. Figure 1 shows a configuration example in which the present invention is applied to a system installed at the entrance of a commercial facility or office to measure the body temperature of visitors without contact to prevent infectious diseases. show.

The surface temperature measurement system 1 is a system that measures the surface temperature of an object to be measured using a thermopile type thermal sensor and displays the measurement results. Thermopile-type thermal sensors have a characteristic that measurement values fluctuate depending on the relative position of the sensor and the object to be measured, so there has been a limit to measurement accuracy in the past. Therefore, in this system 1, the "distance between the thermal sensor and the object to be measured in the optical axis direction of the thermal sensor" and the "positional shift amount of the object to be measured with respect to the optical axis of the thermal sensor" are acquired, and the "distance" is The measured value of the thermal sensor is corrected based on both the "positional deviation amount". By adopting such a configuration, it is possible to realize a system that can accurately measure the surface temperature of the object to be measured while using a thermopile type sensor. Therefore, it is possible to provide a low-cost system, and it is possible to promote its introduction to various places such as commercial facilities and offices, and its spread to developing countries.

In addition, although FIG. 1 illustrates human body temperature measurement, the scope of application of the present invention is not limited to this. For example, the system according to the present invention can be applied to measuring the surface temperature of living things other than humans (such as animals and plants), and measuring the surface temperature of general objects such as machines, buildings, and manufactured products.

<First embodiment>
2 and 3 show a configuration example of the surface temperature measurement system 1 of the first embodiment. FIG. 2 is a perspective view showing the appearance of the surface temperature measurement system 1 of the first embodiment, and FIG. 3 is a block diagram showing the hardware configuration of the surface temperature measurement system 1 of the first embodiment.

The surface temperature measurement system 1 has a thermopile type thermal sensor 10, a camera (image sensor) 11, a processing unit 12, and a display device 13 as main hardware components. The surface temperature measurement system 1 may be designed as a dedicated device, or may be configured by connecting the thermal sensor 10 to a general-purpose computer device (smartphone, tablet, laptop, etc.) equipped with a camera.

The thermal sensor 10 and camera 11 are installed so that their optical axes are parallel (including a state where they are substantially parallel) and have a common field of view. Note that the fields of view of the thermal sensor 10 and the camera 11 do not need to completely match, but only need to overlap at least partially (an area that includes the range where the measurement target object can exist). In the following description, the optical axis direction of the thermal sensor 10 and camera 11 will be referred to as the z direction (depth direction), and the horizontal and vertical directions in a plane perpendicular to the optical axis will be referred to as the x direction and y direction, respectively.

(thermal sensor)
FIG. 4 is a cross-sectional view schematically showing the configuration of the thermal sensor 10. As shown in FIG.

The thermal sensor 10 is an infrared sensor, and includes an optical system 40 that collects radiant thermal energy (infrared rays) from an object to be measured, a thermopile chip 41 that converts the infrared rays into electrical energy, and a thermopile chip 41 that converts the infrared rays into electrical energy. It has a processing circuit 42 that amplifies and AD converts the voltage signal and outputs temperature data. The thermopile chip 41 is a MEMS element and has a structure in which a plurality of thermocouples are arranged in a two-dimensional array.

In this embodiment, for example, a thermal sensor 10 is used that has viewing angles of 90° and 90° in the x direction and y direction, respectively, and a resolution of 32 pixels x 32 pixels in the x direction x y direction. That is, the temperature data output from the thermal sensor 10 becomes a 32×32 two-dimensional image (an image representing temperature distribution). Note that the viewing angle and resolution may be appropriately designed according to the application of the system 1.

(camera)
The camera 11 is a camera for capturing an optical image. For example, a high-resolution camera using an image sensor such as a CCD or CMOS is used.

(processing unit)
The processing unit 12 is a device that performs overall control of the surface temperature measurement system 1, various signal processing and image processing, output processing of measurement results, and the like. The processing unit 12 may be composed of, for example, a general-purpose processor (such as a CPU) and memory, or may be composed of a dedicated processor (such as an ASIC).

(display device)
The display device 13 is a device for outputting information such as measurement results (ie, the surface temperature of the object to be measured). For example, a liquid crystal display or an organic EL display is used.

(Functional configuration)
FIG. 5 is a block diagram showing the functional configuration of the surface temperature measurement system 1 of the first embodiment.

The surface temperature measurement system 1 includes a temperature data acquisition section 50, an image data acquisition section 51, an object detection section 52, a distance calculation section 53, a positional deviation calculation section 54, a temperature correction section 55, and an information output section 56 as functional configurations. have These functions are provided by processing unit 12. For example, if the processing unit 12 is composed of a general-purpose processor, these functions are realized by the processor reading and executing a program stored in a memory. Note that some of the functions shown in FIG. 5 may be realized by a cloud server.

(Surface temperature measurement method)
FIG. 6 is a flowchart showing the flow of surface temperature measurement processing by the surface temperature measurement system 1 of the first embodiment.

In step S60, the temperature data acquisition unit 50 acquires temperature data from the thermal sensor 10. The temperature data acquired here is, for example, a 32×32 two-dimensional image, and the value of each pixel represents the measured value (surface temperature) of the thermal sensor 10.

In step S61, the image data acquisition unit 51 acquires image data from the camera 11. The image data acquired here is an optical image, for example, a 32×32 two-dimensional RGB image. In this embodiment, in order to simplify the explanation, the field of view and resolution of the thermal sensor 10 and the camera 11 are the same (that is, the pixel (x, y) of the temperature data and the pixel (x, y) of the image data are represent the same point on the object to be measured). If the field of view and/or resolution of the two is different, the image data acquisition unit 51 may perform resolution conversion or cropping of the image data in accordance with the temperature data of the thermal sensor 10.

Note that in FIG. 6, image data is captured after temperature data is captured, but the order of capture may be reversed or may be captured in parallel processing. It is desirable that the temperature data and the image data be data measured at the same time or as close as possible.

In step S62, the object detection unit 52 analyzes the image data and detects the measurement object from the image. For example, when the purpose is to measure a person's body temperature, a method of estimating body temperature from the surface temperature of the face is often implemented. Therefore, a "face" as a measurement target may be detected using object detection processing such as a face detection algorithm or a human body detection algorithm. FIG. 7 shows an example of detecting a face 70. As a detection result of the target object detection unit 52, the coordinates (xc, yc) and size (w, h) of the center point of the detection frame 71 surrounding the face 70 are output. Coordinates (xc, yc) represent the detected position on the image, and size (w, h) represents the detected size on the image. FIG. 8 is a diagram illustrating the positional relationship between the optical axis and field of view of the sensor and the measurement target.

In step S63, the distance calculation unit 53 calculates the distance in the optical axis direction from the thermal sensor 10 to the object to be measured based on the detected size (w, h) on the image. The size of the subject on the image depends on the distance between the camera 11 and the subject; the closer the distance is, the larger the size on the image is, and conversely, the farther the distance is, the smaller the size on the image is. Therefore, if the actual size of the subject and the specifications of the optical system and image sensor of the camera 11 are known, the detected size on the image can be converted into the distance between the camera 11 and the subject. Since the camera 11 and the thermal sensor 10 are arranged close to each other, the distance between the camera 11 and the subject may be regarded as the distance between the thermal sensor 10 and the subject (if the distance between the thermal sensor 10 and the camera 11 is If the installation positions are far apart, the distance between the thermal sensor 10 and the subject may be calculated from the distance between the camera 11 and the subject based on the relative positional relationship between the thermal sensor 10 and camera 11.) .

For example, if an LUT or function for converting the size on the image into a distance is prepared in advance, assuming the average size of an adult's face, the distance calculation unit 53 can calculate the size obtained in step S62. The distance can be easily determined by simply substituting the detected size (w, h) into an LUT or function. Note that because the average face size differs depending on gender (male/female) and age (child/adult), multiple types of LUTs or functions are prepared and the object detection unit 52 estimates gender and age. The LUT or function used for distance calculation may be changed based on the estimation result. This makes it possible to improve the accuracy of distance estimation.

In step S64, the positional deviation calculation unit 54 calculates the amount of positional deviation of the measurement target with respect to the optical axis of the thermal sensor 10 based on the detected position (xc, yc) on the image. If the subject is on the optical axis, the subject is at the center of the image, and the farther the subject is from the optical axis, the further away from the center of the image. Therefore, if the specifications of the optical system and image sensor of the camera 11 are known, it is possible to convert the detected position on the image into the amount of positional deviation with respect to the optical axis. Since the camera 11 and the thermal sensor 10 are arranged close to each other and their optical axes are parallel to each other, the amount of displacement of the camera 11 with respect to the optical axis may be regarded as the amount of displacement of the thermal sensor 10 with respect to the optical axis. (If the thermal sensor 10 and camera 11 are installed far apart or their optical axes are not parallel, the amount of positional deviation of the camera 11 with respect to the optical axis will be determined based on the relative positional relationship between the thermal sensor 10 and camera 11. The amount of positional deviation of the thermal sensor 10 with respect to the optical axis can be calculated from . Note that if the distance between the thermal sensor 10 and the object to be measured is required to calculate the amount of positional deviation, the calculation result of the distance calculation section 53 may be used.

In step S65, the temperature correction unit 55 corrects the temperature data obtained in step S60 based on the "distance" calculated in step S63 and the "positional deviation amount" calculated in step S64. As a characteristic of a thermopile type thermal sensor, the measured value tends to decrease as the distance from the thermal sensor to the object to be measured increases. Furthermore, there is a tendency that the larger the amount of positional deviation of the thermal sensor from the optical axis, the smaller the measured value. Therefore, the temperature correction unit 55 may correct the temperature data (measured value) of the thermal sensor 10 so that the longer the distance, the higher the temperature, and the larger the amount of positional deviation, the higher the temperature.

For example, the temperature correction unit 55 includes an LUT in which a "correction amount" for a combination of "distance" and "positional deviation amount" is defined, a function that calculates a "correction amount" from "distance" and "positional deviation amount", etc. The temperature data may be corrected using the following. The LUT and the function may be created from the results of measurement experiments using the thermal sensor 10, or may be created theoretically based on the characteristics of the thermal sensor 10.

In step S66, the information output unit 56 outputs the corrected temperature data to the display device 13. In the output example shown in FIG. 9A, the detection result (detection frame) in step S62 and the temperature corrected in step S65 are displayed superimposed on the image data captured in step S61. When a plurality of measurement objects (faces in this example) are detected from the image as shown in FIG. 9B, the detection frame and correction temperature may be displayed for each measurement object.

According to the configuration described above, since the measured value of the thermal sensor 10 is corrected based on both the "distance" and "positional deviation amount" of the measurement target, it is possible to obtain measurement results with higher accuracy than conventional methods. .

<Second embodiment>
In the second embodiment, the measured value of the thermal sensor 10 is corrected using AI (artificial intelligence). Below, configurations that are different from the first embodiment will be mainly described, and descriptions of configurations that are common to the first embodiment will be omitted.

FIG. 10 shows an example of the functional configuration of the surface temperature measurement system 1 of the second embodiment. The surface temperature measurement system 1 of this embodiment has a temperature data acquisition section 50, an image data acquisition section 51, a temperature correction section 100, and an information output section 56 as functional configurations. The only difference from the first embodiment is the configuration of the temperature correction section 100.

The temperature correction unit 100 is a trained machine that has been machine-learned to output corrected temperature data corrected according to the "distance" and "positional deviation amount" of the measurement target when inputted with image data and temperature data. It has a model 101. Any machine learning method or model structure may be used; for example, a deep neural network may be used. In the learning phase, machine learning is performed using learning data consisting of "image data taken by a camera," "temperature data consisting of measured values output from a thermal sensor," and "true value temperature data." good.

FIG. 11 is a flowchart showing the flow of surface temperature measurement processing by the surface temperature measurement system 1 of the second embodiment. The processing in steps S60 to S61 is the same as in the first embodiment (FIG. 6). In step S110, the temperature correction unit 100 inputs the temperature data obtained in step S60 and the image data obtained in step S61 to the learned model to obtain corrected temperature data. The process in step S66 is the same as in the first embodiment (FIG. 6).

<Third embodiment>
In the first embodiment, the "distance" and "positional deviation" of the object to be measured are determined by analyzing the optical image obtained from the camera 11, whereas in the third embodiment, the output from the thermal sensor 10 is The "distance" and "positional deviation amount" of the object to be measured are determined by analyzing the temperature distribution image. Below, configurations that are different from the first embodiment will be mainly described, and descriptions of configurations that are common to the first embodiment will be omitted.

12 and 13 show a configuration example of a surface temperature measurement system 1 according to the third embodiment. FIG. 12 is a block diagram showing the hardware configuration of the surface temperature measurement system 1 of the third embodiment, and FIG. 13 is a block diagram showing the functional configuration of the surface temperature measurement system 1 of the third embodiment.

The surface temperature measurement system 1 of this embodiment includes a thermopile type thermal sensor 10, a processing unit 12, and a display device 13 as main hardware components. The surface temperature measurement system 1 also includes a temperature data acquisition section 50, a target object detection section 130, a distance calculation section 131, a positional deviation calculation section 132, a temperature correction section 55, and an information output section 56 as functional configurations. The difference from the first embodiment is that a camera is not an essential component and that temperature data (temperature distribution image) is used to calculate distance and positional deviation.

FIG. 14 is a flowchart showing the flow of surface temperature measurement processing by the surface temperature measurement system 1 of the third embodiment. The process in step S60 is the same as in the first embodiment (FIG. 6). In step S140, the object detection unit 130 analyzes the temperature distribution image acquired in step S60, and detects the measurement object from the image. In general, faces and human body parts have higher temperatures than the background, so it is possible to detect faces and human bodies from temperature distribution images by using algorithms such as silhouette detection, for example. In step S141, the distance calculation unit 131 calculates the distance in the optical axis direction from the thermal sensor 10 to the object to be measured based on the detected size on the temperature distribution image. In step S142, the positional deviation calculation unit 132 calculates the amount of positional deviation based on the detected position on the temperature distribution image. Note that the specific method for calculating the distance and positional shift amount may be the same as the method described in the first embodiment. The processing in steps S65 to S66 is the same as in the first embodiment.

The configuration of this embodiment also enables temperature measurement with higher accuracy than before. Moreover, in the configuration of this embodiment, since a camera is not an essential component, the device configuration is simpler than that of the first embodiment, and costs can be reduced.

<Fourth embodiment>
In the first embodiment, the "distance" and "positional deviation amount" of the measurement target were obtained by analyzing the optical image obtained from the camera 11, whereas in the fourth embodiment, the "distance" and "positional deviation amount" of the measurement target were obtained by The "distance" and "positional deviation amount" of the object to be measured are determined by analyzing the distance image. Below, configurations that are different from the first embodiment will be mainly described, and descriptions of configurations that are common to the first embodiment will be omitted.

FIGS. 15 and 16 show a configuration example of the surface temperature measurement system 1 of the fourth embodiment. FIG. 15 is a block diagram showing the hardware configuration of the surface temperature measurement system 1 of the fourth embodiment, and FIG. 16 is a block diagram showing the functional configuration of the surface temperature measurement system 1 of the fourth embodiment.

The surface temperature measurement system 1 of this embodiment includes a thermopile type thermal sensor 10, a TOF sensor 150, a processing unit 12, and a display device 13 as main hardware configurations. The surface temperature measurement system 1 also includes a temperature data acquisition section 50, a distance image acquisition section 160, a target object detection section 161, a distance calculation section 162, a positional deviation calculation section 163, a temperature correction section 55, and an information output section. It has 56. The difference from the first embodiment is that a camera is not an essential component, a TOF sensor is included, and a distance image is used to calculate distance and positional deviation.

FIG. 17 is a flowchart showing the flow of surface temperature measurement processing by the surface temperature measurement system 1 of the fourth embodiment. The process in step S60 is the same as in the first embodiment (FIG. 6). In step S170, the distance image acquisition unit 160 acquires distance image data from the TOF sensor 150. The distance image data acquired here is, for example, a 1024×1024 two-dimensional image, and the value of each pixel represents the distance from the TOF sensor 150 to the subject. As the TOF sensor 150, a sensor using a direct TOF method or a sensor using an indirect TOF method may be used. In step S171, the target object detection unit 161 analyzes the distance image acquired in step S170, and detects a measurement target from the image. For example, by using algorithms such as silhouette detection and point cloud clustering, faces and human bodies can be detected from distance images. In step S172, the distance calculation unit 162 extracts the distance in the optical axis direction from the thermal sensor 10 to the object to be measured from the distance image. In step S173, the positional deviation calculation unit 163 calculates the amount of positional deviation based on the detected position on the distance image. Note that the specific method for calculating the amount of positional deviation may be the same as the method described in the first embodiment. The processing in steps S65 to S66 is the same as in the first embodiment.

The configuration of this embodiment also enables temperature measurement with higher accuracy than before. Moreover, since the configuration of this embodiment uses a TOF sensor, the distance between the sensor and the object to be measured can be accurately known, and the accuracy of temperature correction based on distance can be improved.

<Others>
The above embodiments are merely illustrative examples of configurations of the present invention. The present invention is not limited to the above-described specific form, and various modifications can be made within the scope of the technical idea. The configurations and processes of the embodiments described above may be combined as appropriate. For example, the trained model of the second embodiment may be applied to the configuration of the third embodiment. In this case, when the temperature distribution image output from the thermal sensor is input to the learned model, it is preferable that a corrected temperature distribution image corrected according to the distance and positional shift amount of the measurement target is output. Furthermore, the trained model of the second embodiment may be applied to the configuration of the fourth embodiment. In this case, when the temperature data output from the thermal sensor and the distance image data output from the TOF sensor are input into the trained model, corrected temperature data that has been corrected according to the distance and positional deviation of the measurement target is output. It would be good if it were done.

<Additional note 1>
(1) A thermopile-type thermal sensor (10) that measures the surface temperature of the object to be measured;
Based on the distance between the thermal sensor (10) and the object to be measured in the optical axis direction of the thermal sensor (10), and the amount of positional deviation of the object to be measured with respect to the optical axis of the thermal sensor (10). , a processing unit (12) that corrects the measured value of the thermal sensor (10);
A surface temperature measurement system (1) comprising:

(2) a step (S60) of measuring the surface temperature of the object to be measured using a thermopile type thermal sensor (10);
Based on the distance between the thermal sensor (10) and the object to be measured in the optical axis direction of the thermal sensor (10), and the amount of positional deviation of the object to be measured with respect to the optical axis of the thermal sensor (10). , a step (S65) of correcting the measured value of the thermal sensor (10);
A surface temperature measuring method characterized by comprising:

1: Surface temperature measurement system 10: Thermal sensor 11: Camera 12: Processing unit 13: Display device 40: Optical system 41: Thermopile chip 42: Processing circuit 70: Face 71: Detection frame 150: TOF sensor

Claims (13)

A thermopile-type thermal sensor that measures the surface temperature of the object to be measured,
Correcting the measured value of the thermal sensor based on the distance between the thermal sensor and the object to be measured in the optical axis direction of the thermal sensor and the amount of positional deviation of the object to be measured with respect to the optical axis of the thermal sensor. a processing unit to
A surface temperature measurement system comprising: further comprising a camera having a common field of view with the thermal sensor,
2. The surface temperature measurement system according to claim 1, wherein the processing unit determines the amount of positional deviation of the measurement object with respect to the optical axis by analyzing an image taken by the camera. 2. The processing unit detects the measurement object from the image by object detection processing, and calculates a positional shift amount of the measurement object with respect to the optical axis based on a detected position on the image. 2. The surface temperature measurement system according to 2. The surface temperature measurement system according to claim 2 or 3, wherein the processing unit calculates the distance to the measurement object by analyzing an image taken by the camera. The surface temperature according to claim 4, wherein the processing unit detects the measurement object from the image by object detection processing, and calculates the distance of the measurement object based on the detected size on the image. measurement system. further comprising a camera having a common field of view with the thermal sensor,
The processing unit is an AI that has been machine learned to perform correction according to the distance and positional deviation amount of the measurement target when inputted with an image taken by the camera and a measurement value output from the thermal sensor. 2. The surface temperature measurement system according to claim 1, wherein the measured value of the thermal sensor is corrected using the following. 2. The surface temperature measurement system according to claim 1, wherein the processing unit determines the amount of positional deviation of the measurement object with respect to the optical axis by analyzing a temperature distribution image output from the thermal sensor. The processing unit detects the measurement target from the temperature distribution image by object detection processing, and determines a positional shift amount of the measurement target with respect to the optical axis based on the detected position on the temperature distribution image. The surface temperature measurement system according to claim 7. The surface temperature measurement system according to claim 7 or 8, wherein the processing unit calculates the distance to the measurement object by analyzing the temperature distribution image. 10. The processing unit according to claim 9, wherein the processing unit detects the measurement object from the temperature distribution image by object detection processing, and calculates the distance of the measurement object based on the detected size on the temperature distribution image. Surface temperature measurement system as described. The processing unit is machine-learned to perform correction according to the distance and positional deviation amount of the measurement target when the temperature distribution image output from the thermal sensor is input.
The surface temperature measurement system according to claim 1, wherein the measured value of the thermal sensor is corrected using I. Further comprising a TOF sensor having a common field of view with the thermal sensor,
2. The processing unit calculates a distance of the object to be measured and an amount of positional deviation of the object to be measured with respect to the optical axis by analyzing a distance image obtained by the TOF sensor. The surface temperature measurement system described in . a step of measuring the surface temperature of the object to be measured using a thermopile type thermal sensor;
Correcting the measured value of the thermal sensor based on the distance between the thermal sensor and the object to be measured in the optical axis direction of the thermal sensor and the amount of positional deviation of the object to be measured with respect to the optical axis of the thermal sensor. the step of
A surface temperature measuring method characterized by comprising:

Images