TWI810635B - Thermal image-based temperature measurement calibration method and thermal image device - Google Patents

Abstract

A thermal image-based temperature measurement calibration method applicable to a thermal image device is provided. The method includes a recording stage, a processing stage and a calibration state. During the recording stage, the thermal image device records a monitored environment to generate a measured thermal image. During the processing stage, a processor performs an operation according to the measured thermal image to generate a target information, wherein the target information corresponds to a target in the monitored environment, and the target information includes a target image block and a target measured temperature corresponding to the target image block. During the calibration stage, the processor obtains a distance compensation value according to a pixel number of the target image block, and the processor performs a calibration operation to the target measured temperature at least according to the distance compensation value to generate a calibrated temperature value corresponding to the target.

Description

Thermal image-based temperature measurement and correction method and thermal image device

The invention relates to temperature measurement, in particular to a thermal image-based temperature measurement correction method and a thermal image device.

The novel coronavirus (COVID-19) has become a highly contagious and fatal virus in recent years, and the common symptom of patients is fever. In order to avoid the spread of the epidemic, temperature control is required in medical institutions and public places. The speed of forehead temperature measurement by manpower is slow, which will easily lead to queuing at the entrance and exit, and greatly increase the labor cost of detection. In addition, common body temperature measurement tools such as forehead thermometers or ear thermometers cannot perform rapid non-contact measurements due to the need to touch the human body, and may even increase the risk of infection due to the close distance between the measurer and the subject .

Although non-contact temperature measurement tools such as thermal imaging devices can provide rapid detection and reduce labor costs, the accuracy at this stage is still not ideal. For example, when the distance between the human body and the thermal imaging device is different, the measured temperature drop is extremely large. In addition, in a multi-person environment, the thermal image generated by the thermal imaging device has multiple heat sources, and it is difficult to accurately identify which heat sources belong to human body temperature. Accurately detect body temperature when a large number of people flow.

The present invention proposes a thermal imaging-based temperature measurement and correction method and thermal imaging device, which can quickly measure the body temperature of multiple people in a multi-person environment, and quickly locate people with abnormal temperatures. The invention adopts non-contact measurement and has the characteristics of high precision and low cost.

According to an embodiment of the present invention, a thermal image-based temperature measurement and correction method is suitable for a thermal image device, the thermal image device includes a thermal image camera and a processor, and the thermal image-based temperature measurement and correction method includes : In the shooting stage, the thermal imaging camera shoots a monitoring environment to generate a measurement thermal image; in the processing stage, the processor performs calculations based on the measurement thermal image to obtain a target information, and the target information corresponds to the monitoring environment In a measurement target, the target information includes a target image block and a target measurement temperature corresponding to the target image block; A distance compensation value, and the processor at least performs a correction operation on the target measurement temperature according to the distance compensation value to generate a correction temperature value corresponding to the measurement target.

According to an embodiment of the present invention, a thermal image-based temperature measurement and correction method, wherein the thermal image device further includes a temperature reference device; the photographing stage of the temperature measurement and correction method further includes: performing heating with the temperature reference device to a designated reference temperature; the processing stage further includes: using the processor to perform calculations based on the measured thermal image to obtain a reference information corresponding to the temperature reference device, the reference information includes a reference image block and corresponding to the a reference measurement temperature of the reference image block; and the calibration stage further includes: calculating an environmental compensation value by the processor according to the designated reference temperature and the reference measurement temperature, wherein the correction temperature value is based on the processor The distance compensation value and the environment compensation value are generated by performing a correction operation on the target measured temperature.

According to a thermal image-based temperature measurement correction method according to an embodiment of the present invention, the processor performs summing of the target measurement temperature, the environment compensation value and the distance compensation value to generate the corrected temperature value.

According to an embodiment of the present invention, a temperature measurement and correction method based on a thermal image, wherein the temperature reference device has a thermal radiation sheet for maintaining the specified reference temperature and a heater capable of heating the thermal radiation sheet, and in the In the processing stage, using the processor to perform calculations based on the measured thermal image to obtain the reference information includes: using the processor to perform calculations on the measured thermal image to obtain at least one candidate reference block of the measured thermal image, wherein The difference between the temperature of each candidate reference block and the specified reference temperature is less than or equal to a threshold value; the processor performs a comparison according to the shape of the at least one candidate reference block and the shape of the heat radiation sheet to determine the One of the at least one candidate reference block is the reference image block; and the reference measurement temperature is calculated by the processor according to the pixel values of the reference image block.

According to an embodiment of the present invention, a thermal image-based temperature measurement and correction method, wherein the processor further stores a distance compensation table, and the distance compensation table records the difference between a plurality of compared pixel quantity values and a plurality of preset distance compensation values The corresponding relationship among them, and obtaining the distance compensation value by the processor according to the number of pixels of the target image block includes: using the processor to search the distance compensation table according to the number of pixels of the target image block to determine these One of the preset distance compensation values is the distance compensation value.

According to an embodiment of the present invention, the thermal image-based temperature measurement and correction method, wherein the processor performs summing of the target measured temperature and the distance compensation value to obtain the corrected temperature value.

According to an embodiment of the present invention, a temperature measurement and correction method based on thermal images further includes a parameter setting stage, the parameter setting stage includes: heating the temperature reference device to a specified test temperature, and the temperature reference device and the thermal The distance between the image cameras is a preset distance; the thermal image camera is used to shoot the temperature reference device to generate a test thermal image; the processor is used to perform calculations based on the test thermal image to obtain test reference information corresponding to the temperature reference device , the test reference information includes a reference image block and a reference measurement temperature; and the distance compensation table is updated by the processor according to the test reference information, wherein the number of pixels in the reference image block of the test reference information is the One of the comparison pixel quantity values, the difference between the reference measurement temperature and the specified test temperature is one of the preset distance compensation values.

According to an embodiment of the present invention, a method for temperature measurement and correction based on thermal imaging, wherein the thermal imaging device further includes a visible light camera, and the shooting stage further includes using the visible light camera to shoot the monitoring environment to generate an optical image, the processing The stage further includes using the processor to perform operations according to the thermal image, the optical image and an object recognition model to determine the target image block.

According to a thermal image-based temperature measurement and correction method according to an embodiment of the present invention, using the processor to obtain the distance compensation value according to the number of pixels of the target image block includes: using the processor to obtain the distance compensation value according to the target image block An operation is performed on the number of pixels and a distance compensation function to obtain the distance compensation value, wherein the distance compensation function represents a fitting curve of a plurality of compared pixel number values and a plurality of preset distance compensation values.

According to an embodiment of the present invention, a method for temperature measurement and correction based on thermal images, wherein the processor performs summing of the target measured temperature and the distance compensation value to obtain the corrected temperature value.

According to an embodiment of the present invention, a thermal image-based temperature measurement and correction method, wherein the thermal image device further includes a temperature reference device; and the temperature measurement and correction method further includes a parameter setting stage, the parameter setting stage includes: Heating the temperature reference device to a specified test temperature; adjusting the distance between the temperature reference device and the thermal imaging camera to a first preset distance; photographing the temperature reference device with the thermal imaging camera to generate a first test thermal image ; using the processor to perform calculations based on the first test thermal image to obtain a first reference information, the first reference information corresponds to the temperature reference device and a first preset distance and includes a first reference image block and a the first reference measurement temperature; calculate a first preset distance compensation value with the processor, the first preset distance compensation value is the difference between the first reference measurement temperature and the specified test temperature; adjust the temperature reference The distance between the device and the thermal image camera is a second preset distance; the temperature reference device is photographed by the thermal image camera to generate a second test thermal image; the processor is used to perform calculations based on the second test thermal image to obtain A second reference information, the second reference information corresponds to the temperature reference device and a second preset distance and includes a second reference image block and a second reference measurement temperature; a second prediction is calculated by the processor Set a distance compensation value, the second preset distance compensation value is the difference between the second reference measurement temperature and the designated test temperature; A curve fitting operation is performed with reference to the number of pixels of the image block, the first preset distance compensation value, and the second preset distance compensation value to generate the distance compensation function.

A thermal imaging device according to an embodiment of the present invention includes: a thermal imaging camera that shoots a monitoring environment to generate a measurement thermal image; and a processor that is electrically connected to the thermal imaging camera, and the processor is based on the Perform calculations on measuring thermal images to obtain target information corresponding to a measurement target in the monitoring environment, the target information including a target image block and a target measurement temperature corresponding to the target image block The processor obtains a distance compensation value according to the number of pixels of the target image block; the processor at least performs correction on the target measurement temperature according to the distance compensation value to generate a corrected temperature value corresponding to the measurement target .

A thermal imaging device according to an embodiment of the present invention further includes a temperature reference device electrically connected to the processor, the temperature reference device performs heating to a specified reference temperature, and the processor executes calculations based on the measured thermal image to obtain A reference information corresponding to the temperature reference device, the reference information includes a reference image block and a reference measurement temperature corresponding to the reference image block, the processor calculates an environment according to the specified reference temperature and the reference measurement temperature The compensation value, wherein the corrected temperature value is generated by performing a correction operation on the target measured temperature by the processor according to the distance compensation value and the environment compensation value.

According to a thermal imaging device according to an embodiment of the present invention, the processor performs summing of the target measured temperature, the environment compensation value and the distance compensation value to generate the corrected temperature value.

According to a thermal image device according to an embodiment of the present invention, the temperature reference device has a thermal radiation sheet for maintaining the specified reference temperature and a heater capable of heating the thermal radiation sheet, and the processor executes the measurement of the thermal image operation to obtain at least one candidate reference block of the measured thermal image, wherein the difference between the temperature of each candidate reference block and the specified reference temperature is less than or equal to a threshold value, and the processor is based on the at least one candidate reference block The shape of the block is compared with the shape of the heat radiation sheet, and one of the at least one candidate reference block is determined to be the reference image block; and the processor calculates the reference measurement according to the pixel value of the reference image block temperature.

According to a thermal imaging device according to an embodiment of the present invention, the processor further stores a distance compensation table, and the distance compensation table records the corresponding relationship between a plurality of compared pixel quantity values and a plurality of preset distance compensation values, and The processor searches the distance compensation table according to the number of pixels of the target image block to determine one of the preset distance compensation values as the distance compensation value.

According to a thermal imaging device according to an embodiment of the present invention, the processor performs summing of the target measured temperature and the distance compensation value to obtain the corrected temperature value.

According to a thermal imaging device according to an embodiment of the present invention, before the temperature reference device is heated to the specified test temperature, the temperature reference device is heated to a specified test temperature, and the distance between the temperature reference device and the thermal imaging camera is one The thermal image camera captures the temperature reference device at a preset distance to generate a test thermal image; the processor executes calculations based on the test thermal image to obtain test reference information corresponding to the temperature reference device, and the test reference information includes a reference an image block and a reference measurement temperature; the processor updates the distance compensation table according to the test reference information, wherein the number of pixels in the reference image block of the test reference information is one of the comparison pixel number values , the difference between the reference measurement temperature and the specified test temperature is one of the preset distance compensation values.

A thermal imaging device according to an embodiment of the present invention further includes a visible light camera electrically connected to the processor, the visible light camera captures the monitoring environment to generate an optical image, and the processor bases the thermal image, the optical image and a The object recognition model performs calculations to determine the target image block.

According to a thermal imaging device according to an embodiment of the present invention, the processor executes an operation according to the number of pixels of the target image block and a distance compensation function to obtain the distance compensation value, wherein the distance compensation function represents a plurality of comparison pixels Fitting curves of quantity values and multiple preset distance compensation values.

According to a thermal imaging device according to an embodiment of the present invention, the processor performs summing of the target measured temperature and the distance compensation value to obtain the corrected temperature value.

According to a thermal imaging device according to an embodiment of the present invention, the thermal imaging device further includes a temperature reference device, before the temperature reference device is heated to the specified reference temperature, the temperature reference device is heated to a specified test temperature; and the thermal The imaging device performs the following steps: adjust the distance between the temperature reference device and the thermal imaging camera to be a first preset distance; use the thermal imaging camera to photograph the temperature reference device to generate a first test thermal image; The first test thermal image executes calculations to obtain a first reference information, the first reference information corresponds to the temperature reference device and a first preset distance and includes a first reference image block and a first reference measurement temperature ; use the processor to calculate a first preset distance compensation value, the first preset distance compensation value is the difference between the first reference measurement temperature and the designated test temperature; adjust the temperature reference device and the thermal imaging camera The spacing is a second preset distance; the temperature reference device is photographed by the thermal image camera to generate a second test thermal image; the processor is used to perform calculations based on the second test thermal image to obtain a second reference information, The second reference information corresponds to the temperature reference device and a second preset distance and includes a second reference image block and a second reference measurement temperature; a second preset distance compensation value is calculated by the processor, the The second preset distance compensation value is the difference between the second reference measurement temperature and the specified test temperature; The quantity, the first preset distance compensation value, and the second preset distance compensation value perform a curve fitting operation to generate the distance compensation function.

The above description of the disclosure and the following description of the implementation are used to demonstrate and explain the spirit and principle of the present invention, and provide a further explanation of the patent application scope of the present invention.

The detailed features and characteristics of the present invention are described in detail below in the implementation mode, and its content is enough to enable any person familiar with the relevant art to understand the technical content of the present invention and implement it accordingly, and according to the content disclosed in this specification, the scope of the patent application and the drawings , anyone who is familiar with the related art can easily understand the ideas and features related to the present invention. The following examples are to further describe the concept of the present invention in detail, but not to limit the scope of the present invention in any way.

FIG. 1 is a block diagram of a thermal imaging device 10 according to an embodiment of the present invention. The thermal imaging device 10 includes a thermal imaging camera 12 , a temperature reference device 14 and a processor 30 . The processor 30 is electrically connected to the thermal imaging camera 12 and the temperature reference device 14 . 2 is a schematic diagram of a thermal imaging device 10, a monitoring environment, and a measurement target 70 according to an embodiment of the present invention. The thermal imaging device 10 is suitable for measuring the temperature of the measurement target 70 in the monitoring environment, such as the temperature of people at the entrances and exits of public places. body temperature and correct the measured temperature. The monitoring environment is, for example, the range that can be photographed by the thermal imaging camera 12, such as the area marked with a dotted line in FIG. 2 ,

The thermal image camera 12 captures the monitoring environment to generate a thermal measurement image. The thermal measurement image has a plurality of pixels, and each pixel value represents a temperature value measured at a position corresponding to the pixel. In practice, the measured thermal image can be converted into a grayscale image to view the temperature distribution in the monitoring environment. The thermal imaging camera 12 is, for example, an infrared camera, but the present invention is not limited thereto.

The temperature reference device 14 is set in the monitoring environment. The temperature reference device 14 performs temperature control to maintain a specified reference temperature. The specified reference temperature should be higher than the temperature of the monitoring environment and within the expected temperature range of the measurement target 70 . For example, the normal forehead temperature ranges from 35 degrees Celsius to 37 degrees Celsius, and the specified reference temperature of the temperature reference device 14 can be set at about 36 degrees Celsius. The temperature reference device 14 has a heat radiation sheet that maintains a specified reference temperature and a temperature controller that can perform temperature control on the heat radiation sheet. The temperature controller can be, for example, a heater, but the temperature controller is not limited to the heater, and can also be a The device with cooling function can also achieve the purpose of temperature control.

The temperature reference device 14 and the thermal image camera 12 can adopt a separate design or an integrated design. The difference lies in the distance between the temperature reference device 14 and the thermal image camera 12. In the integrated design, the distance is greater than a preset value. The separation is designed so that the distance is less than the preset value. The separate design is shown in Figure 2, and the measurement thermal image captured by the separate design is shown in Figure 3. The reference image block 141 in Figure 3 corresponds to the temperature reference device 14 captured by the thermal image camera 12, and the target The image block 701 corresponds to the measurement target 70 captured by the thermal imaging camera. The integrated design is shown in FIG. 4 . The temperature reference device 14 is set above the lens of the thermal imaging camera 12 , so the captured thermal measurement image is shown in FIG. 5 , and the reference image block 141 is located above the thermal measurement image. The present invention does not limit the installation position of the temperature reference device 14 in the integrated design, for example, the temperature reference device 14 can be installed at the upper left corner, lower left corner, lower right corner, upper right corner, left, right, or upper of the lens of the thermal imaging camera or below; the measured thermal image and the position of the reference image block 141 captured by the above-mentioned integrated design are shown in FIG. 6 .

The processor 30 performs calculations based on the measured thermal image to obtain reference information and target information. The reference information corresponds to the temperature reference device 14. The reference information includes a reference image block 141 and a reference measurement temperature corresponding to the reference image block 141; the target The information corresponds to the measurement target 70 , and the target information includes the target image block 701 and the target measurement temperature corresponding to the target image block 701 . The processor 30 executes calibration according to the reference information, target information, and specified reference temperature to generate the corrected temperature value of the measurement target 70. The temperature measurement calibration method based on a thermal image according to an embodiment of the present invention will be used to illustrate the calibration temperature generated by the processor 30. value details.

7 is a flow chart of a temperature measurement and correction method based on a thermal image according to an embodiment of the present invention. The method includes a parameter setting stage P0, a photographing stage P1, a processing stage P2, and a correction stage P3.

Before the thermal imaging device 10 of an embodiment of the present invention is put into practical use, the parameter setting stage P0 may be performed multiple times to establish a distance compensation table for subsequent use in correcting the measured temperature. The distance compensation table records correspondences between multiple comparison pixel quantity values and multiple preset distance compensation values. Table 1 below is an example of the distance compensation table. The distance compensation table can include multiple columns, and each column records the preset area (ie: the number of pixels compared), the preset distance and the distance compensation value (ie: the preset distance compensation value ).

Table I Default area A ₀ preset distance D Distance compensation value ∆T _dis 81 pixels 3.0 meters 0.1°C 64 pixels 3.2 meters 0.5°C 49 pixels 3.4 meters 1.0°C

Fig. 8 is a flow chart of establishing a distance compensation table in the parameter setting phase P0. Step S81 is "determining a preset distance between the temperature reference device and the thermal imaging camera", and the preset distance is, for example, 3.0, 3.2 or 3.4 meters as shown in Table 1. Step S82 is "heating to maintain the temperature reference device at the specified test temperature", in detail, the temperature reference device is heated to the specified test temperature, and the specified test temperature should be higher than the temperature of the monitoring environment (such as room temperature) and between the measurement target 70 for the expected temperature range (e.g. the temperature of the human body). Step S83 is "the thermal image camera captures the temperature reference device maintaining the specified test temperature to generate a test thermal image". Step S84 is "the processor executes calculation according to the test thermal image to obtain reference information". The reference information corresponds to the temperature reference device 14 and includes a reference image block 141 and a reference measurement temperature. The reference image block 141 is an image block composed of a plurality of pixels corresponding to the temperature reference device 14 in the test thermal image, and the reference measurement temperature is a temperature value calculated based on a plurality of pixel values of the reference image block 141. The present invention The calculation method of the reference measurement temperature is not limited, for example, the average value (or maximum value) of the pixel values in all the reference image blocks 141 is set as the reference measurement temperature. A first implementation of step S84 is to define a specific location in the test thermal image as the reference image block 141 . For example, in FIG. 5 , the user can designate a specific area above the test thermal image as the reference image block 141 . A second implementation of step S84 is that the processor 30 automatically detects the reference image block 141 in the test thermal image.

FIG. 9 is a flowchart of a second implementation manner of step S84. The temperature reference device 14 has a surface emitting a specified test temperature, and the thermal imaging camera 12 can photograph the surface. Therefore, the processor 30 can detect a plurality of pixels corresponding to the surface in the test thermal image. Step S91 is "the processor obtains each pixel value of the test thermal image", and each pixel value represents the temperature value measured at the position corresponding to the pixel. Step S92 is "the processor marks a plurality of pixels corresponding to the specified test temperature", in other words, the processor 30 judges whether each pixel value of the test thermal image corresponds to the specified test temperature, and if so, marks the pixel, otherwise it does not mark . Step S93 is "the processor judges whether the shape of the marked pixel is the same as the shape of the surface of the temperature reference device", if yes, it means that the processor 30 has found the reference image block 141 in the test thermal image, and the processor 30 records Positions of a plurality of pixels corresponding to the specified test temperature are set as the reference image block 141 .

Please refer to FIG. 8 , step S85 is "update the distance compensation table by the processor", in detail, the processor 30 updates the distance compensation table according to the test reference information. The processor 30 sets the area of the reference image block 141 obtained in step S84 as a preset area, and sets the difference between the reference measurement temperature and the specified test temperature as a preset distance compensation value. The calculation method of the area of the reference image block 141 is, for example, the number of pixels, or the area of the reference image block 141 is calculated according to the shape of the reference image block 141 , which is not limited in the present invention. The reference measurement temperature is usually lower than the specified test temperature, because the airflow in the environment where the temperature reference device 14 is located will take away the radiant heat emitted by the temperature reference device 14 .

In step S85, the processor 30 judges "whether the distance compensation table is completed". For example, the expected measurement distance from the thermal imaging camera 12 is from 3 to 4 meters from the measurement target 70, and the interval of the preset distance is 0.2 meters, then the distance compensation table has 6 columns, and these 6 columns correspond to the preset The distances are 3 meters, 3.2 meters, 3.4 meters, 3.6 meters, 3.8 meters, and 4 meters; the numbers used in the above examples are for illustration but not limitation of the present invention. If the distance compensation table has been completed, continue to execute the shooting phase P1 of FIG. 7; otherwise, return to step S81, select another preset distance, and repeat the process of steps S82-S84.

In practice, the process in FIG. 8 can be adaptively modified, and the establishment of the distance compensation table is changed to the establishment of the distance compensation function. For example, in step S81, multiple preset distances are selected at one time, and multiple test thermal images are taken according to these preset distances in steps S82~S84, and then multiple distance compensation values are calculated by the processor 30, and then based on the above multiple The preset distance and multiple distance compensation values are used to establish the distance compensation function by means of curve fitting, and the curve corresponding to the distance compensation function is shown in FIG. 10 . Specifically, the temperature reference device 14 is heated to a specified test temperature; the distance between the temperature reference device 14 and the thermal imaging camera 12 is adjusted to a first preset distance; the thermal imaging camera 12 shoots the temperature reference device 14 to generate a first test thermal image; The processor 30 executes an operation according to the first test thermal image to obtain first reference information, the reference information corresponds to the temperature reference device and the first preset distance and includes a first reference image block and a first reference measurement temperature; Calculate a first preset distance compensation value with the processor, and the first preset distance compensation value is the difference between the first reference measurement temperature and the specified test temperature; adjust the distance between the temperature reference device 14 and the thermal imaging camera 12 as A second preset distance; the thermal image camera 12 photographs the temperature reference device 14 to generate a second test thermal image; the processor 30 executes calculations based on the second test thermal image to obtain second reference information, and the second reference information corresponds to the temperature reference The device 14 is at a second preset distance and includes a second reference image block and a second reference measurement temperature; the processor 30 calculates a second preset distance compensation value, and the second preset distance compensation value is the second reference measurement temperature The difference with the specified test temperature; and the processor 30 executes according to the number of pixels of the first reference image block, the number of pixels of the second reference image block, the first preset distance compensation value, and the second preset distance compensation value A curve fitting operation to generate a distance compensation function. In one embodiment, the processor 30 performs summing of the target measured temperature and the distance compensation value to obtain the corrected temperature value

In order to improve the measurement accuracy of the thermal imaging camera 12, in the parameter determination stage P0, an additional self-calibration process of the full-plane temperature can be added, as shown in Figure 11: Step S111 is "blackbody furnace initialization", for example, set the blackbody furnace to It is 37 degrees Celsius. Step S112 is "photographing the blackbody furnace with a thermal image camera to generate a test thermal image". Step S113 is "the processor updates the compensation matrix according to the test thermal image". Since the blackbody furnace has a larger surface than the temperature reference device 14, and the thermal radiation is more stable, theoretically, the test thermal image corresponding to the blackbody furnace Each pixel value is the same, for example, both represent 37 degrees Celsius. Therefore, if the processor 30 finds that a certain pixel value is different from the initial temperature value of the blackbody furnace, the difference between the pixel value and the initial temperature value of the blackbody furnace is recorded in the compensation matrix, wherein the elements of the compensation matrix correspond one by one Pixels for testing thermal images. Step S114 is "photographing a temperature reference device with a specified test temperature by the thermal imaging camera", and step S115 is that the processor 30 judges "whether the error between the reference measurement temperature and the specified test temperature is less than a threshold" according to the compensation matrix. If the determination in step S115 is "Yes", it means that the thermal imaging camera 12 has completed the full-plane temperature self-calibration. The updated compensation matrix will be stored in the thermal imaging camera 12 , and will be reused by the thermal imaging camera 12 when performing the shooting phase P1 in FIG. 7 . If the determination in step S115 is "No", then return to step S111 for recalibration.

Please refer to FIG. 7 , the flow from the photographing stage P1 , the processing stage P2 to the correction stage P3 is equivalent to the actual application of the thermal imaging device 10 according to one embodiment of the present invention in a monitoring environment. During the photographing phase P1, the thermal imaging device 10 photographs the monitoring environment to generate a measurement thermal image, and the monitoring environment has the measurement target 70 and the temperature reference device 14 . In the processing stage P2, the processor 30 executes calculations according to the measured thermal image to obtain target information and reference information. The target information corresponds to the measurement target 70 in the monitoring environment, and the target information includes the target image block 701 and the target measurement temperature corresponding to the target image block 701 . The reference information corresponds to the temperature reference device 14 , and the reference information includes the reference image block 141 and the reference measurement temperature corresponding to the reference image block 141 .

In the processing stage P2, the processor 30 executes an operation according to the thermal measurement image to obtain reference information including the following steps: the processor 30 executes an operation on the thermal measurement image to obtain at least one candidate reference block of the thermal measurement image, wherein each The difference between the temperature of a candidate reference block and the specified reference temperature is less than or equal to the threshold value; the processor 30 performs a comparison based on the shape of at least one candidate reference block and the shape of the heat radiation sheet, and determines at least one candidate reference block One of them is the reference image block 141 ; and the processor 30 calculates the reference measurement temperature according to the pixel values of the reference image block 141 . The present invention does not specifically limit the value of the threshold value. In an embodiment, the temperature of the candidate block is, for example, the average value of the temperature values of all pixels in the candidate block.

In the calibration phase P3, the processor 30 executes calibration according to the target information and compensation values (including environmental compensation values and distance compensation values) to generate the corrected temperature value of the measurement target 70. In other words, the processor 30 performs calibration based on at least the distance compensation value The target measurement temperature performs a correction operation to generate a corrected temperature value corresponding to the measurement target 70 . In one embodiment, the processor 30 performs summing of the target measured temperature, the environment compensation value and the distance compensation value to generate the corrected temperature value. In another embodiment, the processor 30 performs summing of the target measured temperature and the distance compensation value to obtain the corrected temperature value.

In an embodiment of the calibration phase P3, the processor 30 calculates the calibration temperature value in the following manner.

(Formula 1)

in To correct the temperature value, To measure the temperature of the target, is the distance compensation value, Compensation value for the environment.

In the processing stage P2, the processor 30 performs calculations based on the measured thermal image to obtain the target measured temperature and the reference measurement temperature, and the processor 30 calculates the difference between the reference measurement temperature minus the specified reference temperature as the environmental compensation value . The present invention uses the relationship between the actual temperature and the measured temperature of the temperature reference device 14 to reflect the influence of the monitoring environment on the measured temperature, and compensates the influence to the corrected temperature value .

In the processing stage P2, the processor 30 executes calculations according to the measured thermal image to obtain the target image block 701 . In one embodiment, the processor 30 according to the area of the target image block 701 Look up the distance compensation table to know the distance between the measurement target 70 and the thermal imaging camera 12 , and further know the distance Corresponding distance compensation value . In another embodiment, the processor 30 takes the area of the target image block 701 Input to the distance compensation function, the output value of the distance compensation function is the distance compensation value . In other words, the processor 30 executes an operation according to the number of pixels of the target image block 701 and the distance compensation function to obtain the distance compensation value, wherein the distance compensation function represents the fitting of multiple comparison pixel quantity values and multiple preset distance compensation values curve.

In an embodiment of the calibration stage P3, the processor 30 calculates the distance compensation value according to and environmental compensation value The sum is used as the compensation value, and the temperature value is corrected measure the temperature of the target and the sum of compensation values.

In other embodiments of the correction phase P3, the compensation value only adopts the distance compensation value and environmental compensation value one of. For example, when the distance compensation table or distance compensation function is not used, the compensation value is the environment compensation value. When the temperature reference device 14 is not used, the compensation value is a distance compensation value.

In practice, when the corrected temperature value calculated by the processor 30 is greater than the preset value, the processor 30 can generate an alarm signal, and the alarm signal is, for example, sent to a display device or a loudspeaker device to notify relevant personnel to deal with it .

In another embodiment of the present invention, the thermal imaging device 10 further includes a visible light camera electrically connected to the processor 30 . In the shooting phase P1, the visible light camera captures the monitoring environment to generate an optical image. In the processing phase P2, the processor 30 can perform calculations based on the optical image, thermal image, and object recognition model to determine the range of the target image block 701. Through the object recognition model The identification accuracy of the target image block 701 can be further improved. FIG. 12 and FIG. 13 are schematic diagrams of a thermal imaging device 10 with a visible light camera 50 adopting a separate design and an integrated design, respectively. The object recognition model can be trained through the steps of testing thermal image collection and object feature marking. In practical application, the thermal image camera 12 inputs the measured thermal image to the object recognition model, and the object recognition model outputs the measured thermal image for subsequent processing after feature selection. The object recognition model can use, for example, the YOLO image recognition model or the human body contour ellipse fitting algorithm, etc., but it is not limited thereto. The object recognition model can also be constructed using historical data training according to the operating environment.

To sum up, the present invention uses the temperature measured by the temperature reference device and the actual temperature to calculate the environmental compensation value, and at the same time uses the thermal image itself to calculate the distance compensation value. With the above two compensation values, the present invention improves the accuracy of temperature correction Spend. In addition, the present invention only uses a low-cost temperature reference device, and can measure the temperature of multiple targets at different distances from the thermal imaging camera without using expensive distance sensors. The invention can quickly measure the body temperature of many people in a multi-person environment, and quickly locate the person with abnormal body temperature. The invention adopts non-contact measurement, which has the advantages of high precision and low cost.

Although the present invention is disclosed by the aforementioned embodiments, they are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, all changes and modifications are within the scope of patent protection of the present invention. For the scope of protection defined by the present invention, please refer to the appended scope of patent application.

10: Thermal imaging device 12: Thermal imaging camera 14: Temperature reference device 141:Reference image block 30: Processor 50:Visible light camera 70: Measurement target 701: target image block P0: parameter setting stage P1: Shooting stage P2: Processing stage P3: Calibration phase S81~S86, S91~S93, S111~S115: steps

FIG. 1 is a block diagram of a thermal imaging device according to an embodiment of the present invention; 2 is a schematic diagram of a thermal imaging device, a monitoring environment and a measurement target according to an embodiment of the present invention; Figure 3 is an example of a thermal image in which the temperature reference device and the thermal imaging camera are designed separately; Fig. 4 is a schematic diagram of an integrated design of a temperature reference device and a thermal imaging camera; Figure 5 is an example of a thermal image with an integrated design of a temperature reference device and a thermal imaging camera; Figure 6 is another example of thermal imaging in which the temperature reference device and thermal imaging camera adopt an integrated design; FIG. 7 is a flowchart of a temperature measurement and correction method based on a thermal image according to an embodiment of the present invention; Fig. 8 is the flowchart of establishing distance compensation table; FIG. 9 is a flowchart of a second implementation of step S84 in FIG. 8; Fig. 10 shows the curve corresponding to the distance compensation function; Fig. 11 is a flow chart of performing full-plane temperature self-calibration to a thermal imaging camera; FIG. 12 is a schematic diagram of a thermal imaging device adopting an integrated design according to another embodiment of the present invention; and FIG. 13 is a schematic diagram of a thermal imaging device adopting a separate design according to another embodiment of the present invention.

10: Thermal imaging device

12: Thermal imaging camera

14: Temperature reference device

70: Measurement target

Claims (22)

A temperature measurement and correction method based on thermal images, suitable for a thermal image device, the thermal image device includes a thermal image camera and a processor, the temperature measurement and correction method based on thermal images includes: a shooting stage, using the thermal image The image camera captures a monitoring environment to generate a measurement thermal image; in the processing stage, the processor executes calculations based on the measurement thermal image to obtain target information, and the target information corresponds to a measurement target in the monitoring environment, The target information includes a target image block and a target measurement temperature corresponding to the target image block; and a correction stage, the processor obtains a distance compensation value according to a pixel quantity of the target image block, and The processor at least performs a correction operation on the target measured temperature according to the distance compensation value to generate a corrected temperature value corresponding to the measured target. The temperature measurement and correction method based on thermal image as described in claim 1, wherein the thermal image device further includes a temperature reference device; the shooting stage of the temperature measurement and correction method further includes: using the temperature reference device to perform heating to a Designate a reference temperature; the processing stage further includes: using the processor to perform calculations based on the measured thermal image to obtain a reference information corresponding to the temperature reference device, the reference information includes a reference image block and a block corresponding to the reference image a reference measurement temperature of the block; and the calibration stage further includes: calculating an environmental compensation value by the processor according to the designated reference temperature and the reference measurement temperature, wherein the correction temperature value is based on the processor according to the distance The compensation value and the environmental compensation value are generated by performing a correction operation on the target measurement temperature. The thermal image-based temperature measurement correction method as described in claim 2, wherein the processor executes summing the target measurement temperature, the environment compensation value and the distance compensation value to generate the corrected temperature value. The method for temperature measurement and correction based on thermal images as described in claim 2, wherein the temperature reference device has a heat radiation sheet that maintains the specified reference temperature and a heater that can heat the heat radiation sheet, and in the processing stage Using the processor to perform calculations based on the measured thermal image to obtain the reference information includes: using the processor to perform calculations on the measured thermal image to obtain at least one candidate reference block of the measured thermal image, each of which The difference between the temperature of the candidate reference block and the specified reference temperature is less than or equal to a threshold value; the processor performs comparison according to the shape of the at least one candidate reference block and the shape of the heat radiation sheet to determine the at least one One of the candidate reference blocks is the reference image block; and the processor calculates the reference measurement temperature according to the pixel values of the reference image block. The temperature measurement and correction method based on thermal image as described in claim 1, wherein the processor also stores a distance compensation table, which records the distance between multiple comparison pixel quantity values and multiple preset distance compensation values Corresponding relationship, and using the processor to obtain the distance compensation value according to the number of pixels of the target image block includes: using the processor to search the distance compensation table according to the number of pixels of the target image block to determine the presets One of the distance compensation values is the distance compensation value. The temperature measurement correction method based on thermal image as described in claim 5, wherein the processor performs summing of the target measured temperature and the distance compensation value to obtain the corrected temperature value. The temperature measurement and correction method based on thermal image as described in claim 5, further includes a parameter setting stage, the parameter setting stage includes: the temperature reference device is heated to a specified test temperature, and the temperature reference device and the thermal image camera The interval is a preset distance; the temperature reference device is photographed by the thermal image camera to generate a test thermal image; the processor is used to perform calculations based on the test thermal image to obtain test reference information corresponding to the temperature reference device, the The test reference information includes a reference image block and a reference measurement temperature; and the distance compensation table is updated by the processor according to the test reference information, wherein the number of pixels in the reference image block of the test reference information is the ratio For one of the pixel quantity values, the difference between the reference measurement temperature and the specified test temperature is one of the preset distance compensation values. The temperature measurement and correction method based on thermal image as described in claim 1, wherein the thermal image device further includes a visible light camera, and the shooting stage further includes shooting the monitoring environment with the visible light camera to generate an optical image, and the processing stage further includes The method includes using the processor to perform calculations according to the thermal image, the optical image and an object recognition model to determine the target image block. The temperature measurement and correction method based on thermal image as described in claim 1, wherein using the processor to obtain the distance compensation value based on the number of pixels of the target image block includes: using the processor to obtain the distance compensation value based on the target image block The number of pixels and a distance compensation function perform an operation to obtain the distance compensation value, wherein the distance compensation function represents a fitting curve of a plurality of compared pixel number values and a plurality of preset distance compensation values. The thermal image-based temperature measurement correction method as described in claim 9, wherein the processor performs summing of the target measured temperature and the distance compensation value to obtain the corrected temperature value. The temperature measurement and correction method based on thermal image as described in claim 9, wherein the thermal image device further includes a temperature reference device; and the temperature measurement and correction method further includes a parameter setting stage, the parameter setting stage includes: using the heating the temperature reference device to a specified test temperature; adjusting the distance between the temperature reference device and the thermal imaging camera to a first preset distance; photographing the temperature reference device with the thermal imaging camera to generate a first test thermal image; The processor executes an operation according to the first test thermal image to obtain a first reference information, the first reference information corresponds to the temperature reference device and a first preset distance and includes a first reference image block and a first reference measurement temperature; calculate a first preset distance compensation value with the processor, the first preset distance compensation value is the difference between the first reference measurement temperature and the specified test temperature; adjust the temperature reference device and The distance between the thermal image cameras is a second preset distance; the thermal image camera is used to shoot the temperature reference device to generate a second test thermal image; the processor is used to perform calculations based on the second test thermal image to obtain a first test thermal image Two reference information, the second reference information corresponds to the temperature reference device and a second preset distance and includes a second reference image block and a second reference measurement temperature; a second preset distance is calculated by the processor Compensation value, the second preset distance compensation value is the difference between the second reference measurement temperature and the designated test temperature; A curve fitting operation is performed on the number of pixels of the block, the first preset distance compensation value, and the second preset distance compensation value to generate the distance compensation function. A thermal imaging device comprising: A thermal image camera, shooting a monitoring environment to generate a measurement thermal image; and a processor, electrically connected to the thermal image camera, the processor executes calculations based on the measurement thermal image to obtain a target information, the target information Corresponding to a measurement target in the monitoring environment, the target information includes a target image block and a target measurement temperature corresponding to the target image block; the processor obtains according to a pixel quantity of the target image block A distance compensation value; the processor corrects the target measured temperature at least according to the distance compensation value to generate a corrected temperature value corresponding to the measurement target. The thermal imaging device as described in claim 12 further includes a temperature reference device electrically connected to the processor, the temperature reference device performs heating to a specified reference temperature, and the processor performs calculations based on the measured thermal image to obtain the corresponding A reference information of the temperature reference device, the reference information includes a reference image block and a reference measurement temperature corresponding to the reference image block, the processor calculates an environmental compensation value according to the specified reference temperature and the reference measurement temperature , wherein the corrected temperature value is generated by performing a correction operation on the target measured temperature by the processor according to the distance compensation value and the environment compensation value. In the thermal imaging device according to claim 12, the processor performs summing of the target measured temperature, the environment compensation value and the distance compensation value to generate the corrected temperature value. As the thermal image device described in claim 12, the temperature reference device has a thermal radiation sheet that maintains the specified reference temperature and a heater that can heat the thermal radiation sheet, and the processor executes an operation on the measured thermal image to Obtaining at least one candidate reference block of the measured thermal image, wherein the difference between the temperature of each candidate reference block and the specified reference temperature is less than or equal to a threshold value, and the processor is based on the at least one candidate reference block shape performs with the shape of the heat radiating sheet comparing and determining one of the at least one candidate reference block as the reference image block; and the processor calculates the reference measurement temperature according to the pixel value of the reference image block. The thermal imaging device as described in claim 12, wherein the processor also stores a distance compensation table, the distance compensation table records the corresponding relationship between a plurality of comparison pixel quantity values and a plurality of preset distance compensation values, and the processing The processor looks up the distance compensation table according to the number of pixels of the target image block to determine one of the preset distance compensation values as the distance compensation value. The thermal imaging device as claimed in claim 12, wherein the processor performs summing of the target measured temperature and the distance compensation value to obtain the corrected temperature value. The thermal imaging device as described in claim 16, before the temperature reference device is heated to the specified test temperature, the temperature reference device is heated to a specified test temperature, and the distance between the temperature reference device and the thermal imaging camera is a preset distance, the thermal image camera captures the temperature reference device to generate a test thermal image; the processor executes calculations based on the test thermal image to obtain test reference information corresponding to the temperature reference device, and the test reference information includes a reference image area block and a reference measurement temperature; the processor updates the distance compensation table according to the test reference information, wherein the pixel quantity of the reference image block of the test reference information is one of the comparison pixel quantity values, the The difference between the reference measurement temperature and the specified test temperature is one of the preset distance compensation values. The thermal imaging device as described in claim 12 further includes a visible light camera electromechanically connected to the processor, the visible light camera photographs the monitoring environment to generate an optical image, and the processor recognizes the thermal image, the optical image and an object The model performs calculations to determine the target image block. As the thermal imaging device described in claim 12, the processor executes an operation according to the number of pixels of the target image block and a distance compensation function to obtain the distance compensation value, wherein the distance compensation function represents a plurality of comparison pixel quantity values and a fitting curve of multiple preset distance compensation values. As for the thermal imaging device described in claim 20, the processor performs summing of the target measured temperature and the distance compensation value to obtain the corrected temperature value. The thermal imaging device as described in claim 20, the thermal imaging device further includes a temperature reference device, and the temperature reference device is heated to a specified test temperature before the temperature reference device is heated to the specified reference temperature; and the thermal imaging device Perform the following steps: adjust the distance between the temperature reference device and the thermal image camera to be a first preset distance; use the thermal image camera to shoot the temperature reference device to generate a first test thermal image; use the processor according to the first A test thermal image performs an operation to obtain a first reference information, the first reference information corresponds to the temperature reference device and a first preset distance and includes a first reference image block and a first reference measurement temperature; The processor calculates a first preset distance compensation value, the first preset distance compensation value is the difference between the first reference measurement temperature and the specified test temperature; adjusts the distance between the temperature reference device and the thermal imaging camera is a second preset distance; use the thermal image camera to shoot the temperature reference device to generate a second test thermal image; use the processor to perform calculations based on the second test thermal image to obtain a second reference information, the first The second reference information corresponds to the temperature reference device and a second preset distance and includes a second reference image block and a second reference measurement temperature; calculating a second preset distance compensation value with the processor, the second preset distance compensation value is the difference between the second reference measurement temperature and the specified test temperature; and using the processor according to the first reference image The pixel quantity of the block, the pixel quantity of the second reference image block, the first preset distance compensation value, and the second preset distance compensation value perform a curve fitting operation to generate the distance compensation function.

Images