JP2009119896A - Face temperature calculation device and face temperature calculation method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enable stable measurement results to be provided considering the face direction of an occupant when measuring the face temperature of the occupant. <P>SOLUTION: A thermal image generation section 21 generates a temperature distribution map IM which indicates temperature distribution within the temperature detection range D of an IR camera 10. A face range F of the occupant P is extracted from the temperature distribution map IM generated by a face range extraction section 22. A face direction determination section 23 determines the face direction of the occupant P. According to this determined face direction, a contribution adjustment section 24 adjusts the contribution of the temperature in the center range FC of the face range F and that of peripheral ranges FR, FR. A face temperature calculation section 25 then calculates the occupant's face temperature from the contribution-adjusted temperature in the center range FC of the face range F and that of peripheral ranges FR, FR. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a face temperature calculation device and a face temperature calculation method for calculating the temperature of an occupant's face.

An example of a vehicle air conditioner that detects the surface temperature of an occupant and controls the air conditioner based on the detected surface temperature is disclosed in Patent Document 1, for example.
JP 2001-199217 A

In the apparatus disclosed in Patent Document 1, in addition to the surface temperature of the occupant's face and the like, the ceiling portion where the surface temperature changes according to the indoor temperature (inside temperature) and the outdoor temperature (outside temperature) The surface temperature of the glass part whose surface temperature changes according to the surface temperature and the sheet part whose surface temperature changes according to the amount of solar radiation is detected using one non-contact temperature sensor.
In addition to the surface temperature of the occupant, the air conditioner is controlled in consideration of the effects of the inside air temperature, the outside air temperature, and the amount of solar radiation so that the temperature inside the vehicle can be controlled appropriately.

However, the non-contact temperature sensor calculates the average temperature in the range visible from the sensor.
For this reason, for example, when measuring the temperature of the occupant's face, the range visible from the sensor changes depending on the orientation of the occupant's face, so the detected face temperature may vary greatly depending on the orientation of the face.
This tendency becomes conspicuous when the skin temperature differs greatly between the forehead portion of the face and the left and right cheek portions due to, for example, the influence of the conditioned air from the air conditioner.

  Therefore, an object of the present invention is to obtain a stable measurement result in consideration of the direction of the occupant's face when measuring the temperature of the occupant's face.

  The present invention generates a thermal image showing a temperature distribution within a detection range of a sensor, extracts an occupant's face area from the generated thermal image, and calculates the occupant's face temperature based on the temperature distribution in the extracted face area The orientation of the occupant's face is determined, and the contribution of the temperature of the central region of the face region and the temperature of the peripheral region when calculating the face temperature is adjusted according to the determined face orientation. .

  According to the present invention, the calculated occupant face temperature does not vary greatly depending on the orientation of the occupant's face.

Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.
FIG. 1 is a block diagram illustrating a configuration of a face temperature calculation apparatus according to an embodiment.
FIG. 2 is a diagram for explaining an installation example of an IR camera, and FIG. 3 is a diagram for explaining a temperature detection range of the IR camera and a temperature distribution map generated based on the output of the IR camera.

  The face temperature calculation apparatus according to the embodiment includes an IR camera 10 and a control unit 20, measures the skin temperature (face temperature) of the occupant's face, and outputs the measurement result to the air conditioner 30.

As shown in FIG. 2, the IR camera 10 is provided on the upper part of the steering column in the vehicle compartment with the sensor surface facing the face of the passenger P of the vehicle V.
The IR camera 10 is an infrared sensor that detects infrared rays radiated from the occupant P as a detection target in a non-contact manner and outputs an electrical signal corresponding to the detected amount of infrared rays.

As shown in FIG. 3 (a), the temperature detection range D of the IR camera 10 is set so as to include the face of the occupant P and its surrounding area, and the temperature detection range D is partitioned in a grid pattern, A plurality of rectangular detection areas S are set.
A plurality of temperature detection elements are arranged in a matrix on the sensor surface of the IR camera 10, and each temperature detection element measures the temperature of a predetermined detection region S within the temperature detection range D, and performs such detection. The average temperature of the region S is calculated.
The IR camera 10 collects the calculation results of the temperature detection elements and outputs them to the control unit 20 as thermal image data indicating the temperature distribution within the temperature detection range D.
Here, the thermal image data output from the IR camera 10 is data defining a temperature distribution map (thermal image) IM indicating the temperature distribution within the temperature detection range D as shown in FIG.

  The control unit 20 includes a thermal image generation unit 21, a face region extraction unit 22, a face orientation determination unit 23, a contribution degree adjustment unit 24, and a face temperature calculation unit 25.

The thermal image generation unit 21 generates a temperature distribution map IM within the temperature detection range D of the IR camera 10 based on the thermal image data input from the IR camera 10.
The face area extraction unit 22 extracts an occupant's face area (face area) from the temperature distribution map.

  The face orientation determination unit 23 identifies the center (face center) of the occupant's face in the left-right direction based on the temperature distribution in the face area, and based on the position of the identified face center in the face area, Determine the direction.

  The contribution adjustment unit 24 determines the contribution of the temperature of the central region of the face region and the temperature of the peripheral region used in calculating the occupant's face temperature in the subsequent face temperature calculation unit 25 according to the orientation of the occupant's face. Adjust.

  The face temperature calculation unit 25 calculates the occupant's face temperature based on the temperature distribution in the face region. In addition, when the contribution degree adjustment unit 24 adjusts the contribution degree between the temperature of the central area of the face area and the temperature of the peripheral area, the face temperature of the occupant is calculated by reflecting the adjusted contribution degree.

  Each functional block of the thermal image generation unit 21, the face region extraction unit 22, the face orientation determination unit 23, the contribution degree adjustment unit 24, and the face temperature calculation unit 25 is configured by a well-known microcomputer. This is realized by a CPU (not shown) included in the calculation device executing a program stored in a ROM or memory (not shown).

FIG. 4 is a flowchart for explaining processing in the face temperature calculation apparatus according to the embodiment.
Processing in the face temperature calculation apparatus according to the embodiment will be described.
When the power of the face temperature calculation device is turned on, in step 100, the face orientation determining unit 23 initializes counter and flag values held in a memory (not shown).
Here, a horizontal counter C, an update flag F1, and a face direction flag F2, which will be described later, are initialized, and their values are set to “0”.

In step 200, the thermal image generation unit 21 acquires thermal image data output from the IR camera 10 and generates a temperature distribution map indicating the temperature distribution within the temperature detection range D of the IR camera 10 (((3) in FIG. 3). b)).
In this temperature distribution map, among the detection areas S set by dividing the temperature detection range D shown in FIG. 3A in a lattice shape, a detection area having a higher temperature is displayed with a higher density value. Here, in the figure, it means that the temperature of the detection region indicated by the symbol a is the highest, and the temperature decreases as the detection region indicated by the symbol d passes through the detection regions indicated by the symbols b and c.

In step 300, the face area extraction unit 22 extracts the face area of the occupant P based on the temperature distribution within the temperature detection range D of the IR camera 10.
Specifically, the face area extraction unit 22 narrows down a range in which the occupant's face is predicted based on the attachment position and angle of the IR camera 10, and then, for example, a predetermined value such as 33 ± 3 ° C. A detection area indicating a temperature within the temperature range is searched. Then, a predetermined range in which the detected detection areas exist in a lump is extracted as the occupant face area F.

  For example, in FIG. 3B, when the temperature of the detection area indicated by the reference sign a to the reference sign c is within a range of 33 ± 3 ° C., the range F in which the detection areas indicated by the reference sign a to the reference sign c are gathered. , The face area F of the passenger.

  In step 400, the face orientation determination unit 23 detects the position of the center (face center) in the left-right direction of the face of the occupant P within the face region F.

FIG. 5 is a diagram for explaining detection of the face center in the face area and determination of the face orientation.
The detection of the center of the face is based on the fact that the temperature of the eyes and nose is lower than that of the forehead, and that glasses and sunglasses are measured as low-temperature objects. Is detected, and the face center position is detected based on the detected detection area.
Thereby, for example, the position indicated by the symbol C in FIG. 5A is detected as the position of the face center C in the face area F.
Note that the position of the face center may be estimated using the right / left target property of the face region extracted in step 300.

  In step 500, the face orientation determination unit 23 determines the face orientation based on the position of the face center C in the face region F. Then, based on the determination result, the values of “face orientation flag F2” and “update flag F1” that are different values depending on the face orientation are set.

The determination of the orientation of the face based on the position of the face center C in the face area F is illustrated.
For example, when the face area F shown in FIG. 5A is extracted, the face orientation determination unit 23 obtains the width W in the left-right direction of the extracted face area F. Then, when the face center C is located at a substantially central position (indicated by the symbol X in the figure) in the width direction of the face region F, it is determined that the face direction is “front-facing”.
Further, when the face center C is located within a predetermined range (a range indicated by a symbol Y in (b) of FIG. 5) set based on the center X in the width direction of the face region F, the face orientation is “oblique” Direction.
Furthermore, when the face center is located within the range indicated by the symbol Z in the figure, the face center is located at a position where one of the cheeks of the face of the occupant P is almost invisible from the IR camera 10. In this case, it is determined that the face orientation is “landscape”.

Processing performed by the face orientation determination unit 23 will be described in detail.
FIG. 6 is a flowchart for explaining processing performed by the face orientation determination unit 23.

  When the processing in step 400 is completed, in step 501, the face orientation determination unit 23 specifies the position of the face center in the face area.

In step 502, the face orientation determination unit 23 determines whether or not the face is “landscape”.
Whether or not the face is “landscape” is determined based on whether the position of the face center C in the face region F is within a region near the right or left end of the face region F (a region indicated by a symbol Z in FIG. 5B). Judgment based on whether or not there is.

If it is determined in step 502 that it is “landscape”, in step 503, the face orientation determination unit 23 increments the landscape counter C by “1”.
Here, the value of the sideways counter C indicates the duration time after the face is determined to be “sideways”.

In step 504, the face orientation determination unit 23 compares the value of the landscape counter C with a predetermined time T0 to check whether the value of the landscape counter C is less than the predetermined time T0.
When the horizontal counter C is less than the predetermined time T0 (C <T0), in step 505, the face direction determination unit 23 sets the value of the update flag F1 to “0” and sets the value of the face direction flag F2 to “3”. To "".

Here, when the sideways counter C is less than the predetermined time T0 (C <T0), the time during which the face faces sideways is a short time, and during this short time, the change in the skin temperature of the face Can be judged to hardly occur.
Therefore, in this case, the value of the update flag F1 is set to “0” and the value of the face direction flag F2 is set to “3” in order to hold the result immediately before the face becomes “landscape”.

  Here, the value of the update flag F1 being “0” means that the calculation result of the face temperature is not updated, and the value of the face direction flag F2 being “3” means that the face is “landscape”. This means that the state is a short time.

  Incidentally, the predetermined time T0 is set to a time when the change in the skin temperature of the face hardly occurs, for example, 30 seconds. That is, if the time during which the face is “landscape” is less than 30 seconds, the value of the update flag F1 is set to “0” because the face temperature hardly changes in a very short time of 30 seconds. Thus, the calculation result of the face temperature is not updated.

If the value of the horizontal direction counter C is not less than the predetermined time T0 in step 504, in step 506, the face direction determination unit 23 sets the value of the update flag F1 to “1” and sets the value of the face direction flag F2 to “2”. Set to.
Here, when the horizontal counter C is not less than the predetermined time T0, that is, when the value of the horizontal counter C is equal to or longer than the predetermined time T0 (C ≧ T0), the temperature of the part of the face that cannot be seen from the IR camera 10 cannot be measured. Judge.

Therefore, in this case, the contribution degree adjustment unit 24 described later uses the temperature of the cheek portion visible from the IR camera 10 to supplement the temperature information of the cheek portion on the side not visible from the IR camera 10. Since it is necessary, the value of the update flag is set to “1”, and the value of the face direction flag F2 is set to “2”.
Here, the value of the update flag F1 being “1” means updating the calculation result of the face temperature, and the value of the face direction flag F2 being “2” means that the face is “landscape”. It means that.

  If it is not determined in step 502 to be “landscape”, that is, if it is determined that the face is not “landscape” and both cheeks of the face are visible from the IR camera 10, face orientation is determined in step 507. The determination unit 23 sets the value of the horizontal counter C indicating “time” after the face is determined to be “horizontal” to “0”.

In step 508, the face orientation determination unit 23 determines whether or not the face is “oblique”.
Whether or not the face is “inclined” is determined based on whether or not the position of the face center C in the face area F deviates from the center position X in the left-right direction of the face area F.
Therefore, when the position of the face center C deviates from the center position X, it is determined that the face is “oblique”, and when it is not deviated, it is determined that the face is not “oblique”.

If it is determined in step 508 that the direction is “oblique”, in step 509, the face direction determination unit 23 sets the value of the update flag F1 to “1” and sets the value of the face direction flag F2 to “1”. Set.
On the other hand, if it is determined in step 508 that it is not “oblique”, the face orientation determination unit 23 sets the value of the update flag to “1” and the value of the face orientation flag F2 to “0” in step 510. Set.
Here, the value of the update flag F1 being “1” means updating the calculation result of the face temperature, and the value of the face direction flag F2 being “1” means that the face is “oblique” And the value of the face direction flag F2 being “0” means that the face is “front-facing”.

In this way, the face orientation determining unit 23 determines the face orientation of the occupant based on the face center position C in the face region F. Then, the value of the face direction flag F2 required for the processing in the contribution level adjustment unit 24 in the subsequent stage and the value of the update flag F1 required in the processing in the face temperature calculation unit 25 in the subsequent stage are set.
The set values of “update flag F1” and “face orientation flag F2” are stored in a memory (not shown).

  In step 600, the contribution degree adjustment unit 24 determines the skin temperature (face temperature) of the occupant's face in the face temperature calculation unit 25 described later based on the update flag F1 and the face orientation flag F2 set in the process in step 500. ) Is calculated and the total number of correction elements used.

The process performed by the contribution adjustment unit 24 will be described in detail.
FIG. 7 is a flowchart for explaining processing performed by the contribution degree adjustment unit 24.
When the processing in step 500 is completed, in step 601, the contribution adjustment unit 24 reads the values of “update flag F1” and “face orientation flag F2” set in step 500 from a memory (not shown).

  In step 602, the contribution degree adjustment unit 24 checks whether or not the value of the read update flag F1 is “1”, that is, whether or not the update of the calculation result of the face temperature is valid.

If the value of the update flag F1 is not “1” in step 602, the process ends.
When the value of the update flag F1 is “1” in step 602, in step 603, the contribution degree adjustment unit 24 determines whether or not the value of the face direction flag F2 is “0”, that is, the face is “front-facing”. It is confirmed whether or not.

  When the value of the face orientation flag F2 is “0” in step 603, in step 604, the contribution degree adjustment unit 24 measures the temperature of the right cheek of the occupant's face and the temperature of the left cheek. The same number of temperature detection elements are selected from the temperature detection elements that measure the temperature. The number of selected temperature detection elements is set as the number of correction elements, and the total value of measured values of the selected temperature detection elements is set as the correction element total value.

Here, the process of step 604 will be described in more detail.
FIG. 8 is an example of a temperature distribution map in the case where the face direction is the front direction, and is a diagram for explaining the central region and the peripheral region in the face region, and the range indicated by the symbol FC is the central region of the face ( Forehead region), and the ranges indicated by symbols FL and FR indicate the temperature distribution of the peripheral region (left cheek region and right cheek region) of the face, respectively.

When the temperature distribution map as shown in FIG. 8A is obtained based on the thermal image data, in step 604, the contribution degree adjustment unit 24 uses the same number from the left cheek region FL and the right cheek region FR. The detection areas S are selected one by one (see FIG. 8B), and the number of selected detection areas is set as the number of correction elements.
As described above, a plurality of temperature detection elements are provided on the sensor surface of the IR camera 10, and each temperature detection element is set so as to calculate an average temperature in the determined detection region S. Therefore, for example, the number of detection areas S selected from the left cheek area FL is the same as the number of temperature detection elements selected from the temperature detection elements for measuring the temperature of the left cheek.

  For the same reason, the contribution adjustment unit 24 obtains the total value of the temperatures of the selected detection region S and sets it as the correction element total value.

  Here, for example, when four detection areas are selected from the left and right cheek areas, and the temperature of each selected detection area is 32 ° C., the number of correction elements is “8”, and the total correction element value is selected. The total value of the temperature in the detection area, that is, “254 (32 × 8)”.

  In this way, when the face temperature calculation unit 25 calculates the occupant's face temperature based on the temperature distribution in the face area F, the peripheral area of the face area F (left cheek area FL, right cheek area FR). The left cheek region FL on the one side in the left-right direction of the face region F across the center region FC of the face region F is further made to be greater than the contribution of the temperature of the center region FC. And the contribution of the temperature of the right cheek region FR on the other side are made equal.

  If the value of the face orientation flag F2 is not “0” in step 603, in step 605, the contribution degree adjustment unit 24 determines whether or not the value of the face orientation flag F2 is “1”, that is, the face is “oblique”. It is confirmed whether or not.

  When the value of the face direction flag F2 is “1” in step 605, in step 606, the contribution degree adjustment unit 24 detects the temperature of the right cheek of the occupant's face and the temperature of the left cheek. The temperature detection element that measures the temperature of the cheek with the smaller area visible from the IR camera 10 is selected. Then, based on the selected temperature detection element, the number of correction elements and the total correction element value are obtained.

Here, the processing of step 606 will be described in more detail.
FIG. 9 is an example of a temperature map when the face orientation is oblique. The range indicated by reference character FC indicates the temperature distribution of the center region (forehead region) of the face, and the ranges indicated by reference characters FL and FR are shown in FIG. The temperature distribution of the peripheral area (left cheek area, right cheek area) of each face is shown.

  When a temperature distribution map as shown in FIG. 9A is obtained based on the thermal image data, in step 606, the contribution degree adjustment unit 24 selects one of the left cheek region FL and the right cheek region FR. A detection region is selected from the smaller area in the detection region F, and the number of selected detection regions is set as the number of correction elements. And the total value of the temperature of the selected detection area | region S is calculated | required, and it is set as a correction element total value.

In the case of FIG. 9A, the number of detection areas in the right cheek area FR is smaller than the number of detection areas in the left cheek area FL, and the area occupied by the right cheek area FR in the face area F is left. Since the area is smaller than the cheek area FL, the detection area is selected from the right cheek area FR.
At this time, a detection area at a position away from the center of the face is selected from the detection areas included in the right cheek area FR.
In the case of FIG. 9A, since only two detection areas are included in the right cheek area FR, these two detection areas are located away from the face center.

  In this way, when the face temperature calculation unit 25 calculates the occupant's face temperature based on the temperature distribution in the face area F, the peripheral area of the face area F (left cheek area FL, right cheek area FR). The contribution of the temperature of the cheek region on the side where the range visible from the IR camera 10 is narrower is larger than the contribution of the temperature of the central region FC. It is set to be larger than the contribution of the temperature of the side cheek region.

If the value of the face orientation flag is not “1” in step 605, that is, if the face is “landscape”, in step 607, the contribution degree adjustment unit 24 determines the temperature of the cheek of the passenger on the side visible by the IR camera 10. Select a temperature detection element to measure. Then, based on the selected temperature detection element, the number of correction elements and the total correction element value are obtained.
This is to supplement the temperature information of the cheek on the side not visible with the IR camera 10 using the temperature information of the cheek on the visible side.

Here, the process of step 607 will be described in more detail. FIG. 10 is an example of a temperature map when the face orientation is oblique, and the range indicated by the symbol FC is the center area (forehead area) of the face. The temperature distribution indicates the temperature distribution in the peripheral area (left cheek area) of the face.

When the temperature distribution map as shown in FIG. 10A is obtained based on the thermal image data, in step 607, the contribution degree adjustment unit 24 determines the left cheek region on the side visible with the IR camera 10. A detection area is selected from the detection areas in the FL, and the number of selected detection areas is set as the number of correction elements. And the total value of the temperature of the selected detection area | region S is calculated | required, and it is set as a correction element total value.
At this time, a detection area at a position away from the center of the face is selected from detection areas included in the left cheek area FL. The detection area located outside the face center in the left cheek area FL is selected, and the contribution is adjusted so that the central area FC and the peripheral areas FL and FR in the face area F are adjusted. This is because the contribution can be adjusted without clearly distinguishing.

  In this way, when the face temperature calculation unit 25 calculates the occupant's face temperature based on the temperature distribution in the face area F, the degree of contribution of the temperature in the peripheral area (left cheek area FL) of the face area F However, the temperature of the cheek region on the side not visible from the IR camera 10 is complemented by the temperature of the cheek region on the visible side. .

Here, in the case of FIG. 10A, the right cheek region FR is not visible, but the right cheek region may be slightly seen even if the face orientation is determined to be landscape.
In this case, the detection area is selected from the one having the larger area in the detection area F of the left cheek area FL and the right cheek area FR.

In this way, the contribution adjustment unit 24 determines the number of correction elements and the total correction element value according to the face orientation determined by the face direction determination unit 23.
The determined number of correction elements and the total correction element value are held in a memory (not shown).

  In step 700, the face temperature calculation unit 25 calculates the occupant's face temperature using the number of temperature detection elements and the element total value calculated in step 600.

Processing performed by the face temperature calculation unit 25 will be described.
FIG. 11 is a flowchart for describing processing performed by the face temperature calculation unit 25.
In step 701, the face temperature calculation unit 25 reads the value of the “update flag F1”, the number of correction elements, and the total correction element value set by the processing in steps 500 and 600 from a memory (not shown).

  In step 702, the face temperature calculation unit 25 checks whether or not the value of the read update flag is “1”, that is, whether or not the update of the calculation result of the face temperature is valid.

When the value of the update flag is “1” in step 702, in step 703, the face temperature calculation unit 25 determines the total number of face area elements and the face area elements in the face area F extracted in the processing in step 300 described above. Specifically, the total value is obtained by counting the number of detection areas S in the face area F to obtain the total number of elements in the face area F, and obtaining the total value of the temperatures of the detection areas S in the face area F. The total value.

In step 704, the face temperature calculation unit 25 calculates the face temperature (average face skin temperature) based on the following equation (1).
Face temperature = (total value of all elements in face area + total value of correction elements) / (total number of elements in face area + number of correction elements) (1)

  In step 705, the face temperature calculation unit 25 sets the calculated face temperature to the value of the storage parameter P_Result and stores it in a memory (not shown).

If the value of the update flag is not “1” in step 702, in step 706, the face temperature calculation unit 25 ends the process without changing the value of the saved parameter P_Result stored in the memory (not shown).
That is, when the value of the update flag is “0” instead of “1”, it indicates the face temperature calculated in the past (the face temperature calculated immediately before it is determined that the face orientation is landscape). The saved parameter P_Result is held in the memory as it is.

  In step 800, the face temperature calculation unit 25 outputs the occupant face temperature calculated in the process in step 700 to the air conditioner 3, and then proceeds to the process in step 200. Thereby, the same processing is repeated thereafter.

  Here, the above-described step 200 and the thermal image generation unit 21 correspond to the thermal image generation unit in the invention, and the step 300 and the face area extraction unit 22 correspond to the face area extraction unit in the invention. The face orientation determination unit 23 corresponds to the face orientation determination unit in the invention, the step 600 and the contribution degree adjustment unit 24 correspond to the contribution degree adjustment unit in the invention, and the step 700 and the face temperature calculation unit 25 in the invention. It corresponds to a face temperature calculation unit.

As described above, in this embodiment, the thermal image generation unit 21 that generates the temperature distribution map IM indicating the temperature distribution in the temperature detection range D of the IR camera 10 and the face region F of the occupant P in the temperature distribution map IM. In a face temperature calculation device including a face area extraction unit 22 to be extracted and a face temperature calculation unit 25 to calculate the face temperature of the occupant P based on the temperature distribution in the face area F, the orientation of the occupant P's face is determined. Adjusting the degree of contribution of the temperature of the central area FC of the face area F and the temperatures of the peripheral areas FL and FR when calculating the face temperature of the occupant P based on the face orientation determination unit 23 that performs The contribution degree adjusting unit 24 is provided.
As a result, when calculating the occupant's face temperature, the calculated occupant's face temperature does not vary greatly depending on the occupant's face orientation. It will be obtained.

Further, the contribution adjustment unit 24 is configured to make the contribution of the temperatures of the peripheral areas FL and FR of the face area F larger than the contribution of the temperature of the central area FC.
As a result, the region near the center (center region FC) in the left-right direction of the face region F of the occupant P is regarded as the forehead region, and the peripheral regions FL and FR positioned to the left and right across the center region FC are regarded as cheek regions. Depending on the orientation of the face, the face temperature is calculated after changing the usage ratio of each area so that the area of the central area FC and the peripheral areas FL and FR are equal. A face temperature reflecting nearby temperature information is calculated.
Therefore, when calculating the occupant's face temperature, the calculated occupant's face temperature does not vary greatly depending on the occupant's face orientation, so a stable measurement result considering the occupant's face orientation is obtained. Be able to.
The same effect can be obtained by making the contribution of the temperature of the central area FC of the face area F smaller than the contribution of the temperatures of the peripheral areas FL and FR.

Further, the face orientation determination unit 23 determines the orientation of the face of the occupant P based on the position of the center C in the left-right direction of the face of the occupant P in the face area F, at least one of front, oblique, and sideways. It was set as the structure which judges.
Thus, the face orientation can be determined by a simple operation of detecting the position of the occupant's face center C in the face area F.

Further, when it is determined that the direction of the face of the occupant P is the front direction, the contribution adjustment unit 24 determines the temperature of the peripheral area FL on one side in the left-right direction of the face area F across the center area FC, The contribution of the temperature of the peripheral region FR on the other side is set to be equal.
The temperature of the peripheral area FL on one side in the left-right direction of the face area F across the central area FC means, for example, the temperature of the left cheek, and the temperature of the peripheral area FR on the other side means, for example, the temperature of the right cheek. Therefore, when there is a temperature difference between the left and right cheeks, the temperature information of the left and right cheeks can be uniformly reflected in the calculation of the face temperature.

Further, when it is determined that the orientation of the face of the occupant P is oblique, the contribution degree adjustment unit 24 has a peripheral area FL on one side in the left-right direction of the face area F across the center area FC, and the other side. Of the peripheral area FR in the area of the face area F, and the contribution of the temperature of the peripheral area with the smaller area in the face area F is set high.
As a result, when the occupant is facing obliquely with respect to the IR camera 10 that is a non-contact temperature sensor, the contribution of the cheek on the side with the smaller area is increased, so that the side with the smaller area is calculated when calculating the face temperature. The cheek temperature information can be reflected to the same extent as the cheek temperature information on the larger side.

Further, the contribution degree adjustment unit 24 determines that the face direction of the occupant P is horizontal and if the time determined to be horizontal is equal to or longer than the predetermined time T0, the face area is sandwiched between the central area FC. Of the peripheral region FL on one side in the left-right direction of F and the peripheral region FR on the other side, the contribution of the temperature of the peripheral region with the larger area in the face region F is set high. .
As a result, when the occupant faces the IR camera 10 that is a non-contact temperature sensor for a predetermined time T0 or more, the temperature of the cheek on the opposite side is determined using the cheek temperature information on the IR camera 10 side. Since the information is complemented, the face temperature can be continuously measured even if the passenger remains facing sideways.

Further, the contribution degree adjustment unit 24 determines that the face direction of the occupant P is horizontal when the face direction of the occupant P is determined to be horizontal and the time determined to be horizontal is less than the predetermined time T0. The face temperature calculated immediately before it is determined to be the occupant P face temperature is used.
As a result, when the time when the occupant faces sideways with respect to the IR camera 10 that is a non-contact sensor is a short time that is less than the predetermined time T0, the face temperature calculated immediately before the occupant faces sideways is maintained. Therefore, the face temperature of the occupant P is output to the air conditioner 30 when the occupant turns his face sideways for a short time for safety confirmation or the like. Can be prevented from affecting the control.

  In the above embodiment, for convenience of explanation, the identification of the face area and the subsequent processing have been described using the temperature distribution map IM generated from the thermal image data. However, the temperature data of each detection area in the temperature detection range is the thermal image. Since the data is digitized, actual processing in each part of the control unit 20 is performed based on the digitized temperature data.

In the above embodiment, the case where the temperature detection range D of the IR camera 10 is set so as to include the face of the occupant P and the surrounding area has been described as an example. It is good also as a temperature detection range which can measure surface temperature.
Alternatively, an IR camera 10 may be prepared for each seat provided in the passenger compartment, and the prepared IR camera 10 may measure the surface temperature of each occupant seated in each seat.

In the above embodiment, the orientation of the occupant's face is determined, and the contribution of the temperature of the central region of the facial region and the temperature of the peripheral region when calculating the facial temperature is adjusted according to the determined facial orientation, The present invention has been described as a function of the face temperature calculation device that calculates the face temperature of the occupant P based on the temperature distribution in the face region F after the degree adjustment.
However, the present invention can also be embodied as a face temperature measurement program that causes a computer to execute the functions of the above-described face temperature calculation device, or a medium that stores this program.

It is a block diagram which shows the structure of the face temperature calculation apparatus concerning an Example. It is a figure explaining the example of installation of IR camera. It is a figure explaining the temperature detection range of IR camera 10, and a temperature distribution map. It is a flowchart explaining the process in the face temperature calculation apparatus concerning an Example. It is a figure explaining the determination of the direction of the face based on the position of the face center. It is a flowchart explaining the detail of the process which the face direction determination part 23 performs. It is a flowchart explaining the detail of the process which the contribution degree adjustment part 24 performs. It is a figure which shows a temperature distribution map in case a face is "frontward." It is a figure which shows a temperature distribution map in case a face is "diagonal." It is a figure which shows a temperature distribution map in case a face is "landscape." It is a flowchart explaining the detail of the process which the face temperature calculation part 25 performs.

Explanation of symbols

DESCRIPTION OF SYMBOLS 3 Air conditioner 6 Crew 10 IR camera 20 Control part 21 Thermal image generation part 22 Face area extraction part 23 Face direction determination part 24 Contribution degree adjustment part 25 Face temperature calculation part 30 Air conditioner C Counter D Temperature detection range F1 Update flag F2 flag FL Left cheek area FR Right cheek area FC Center area IM Temperature distribution map

Claims (8)

A sensor,
A thermal image generation unit that generates a thermal image indicating a temperature distribution within a detection range of the sensor;
A face area extraction unit for extracting a passenger's face area from the thermal image;
In a face temperature calculation device comprising a face temperature calculation unit that calculates the face temperature of the occupant based on the temperature distribution in the face region,
A face orientation determination unit that determines the orientation of the occupant's face;
And a contribution degree adjusting unit that adjusts the contribution of the temperature of the central area of the face area and the temperature of the peripheral area when calculating the face temperature of the occupant based on the orientation of the face of the occupant. Facial temperature calculation device. The contribution degree adjustment unit
The contribution of the temperature of the central region is made smaller than the contribution of the temperature of the peripheral region, or the contribution of the temperature of the peripheral region is made larger than the contribution of the temperature of the central region. Item 2. The face temperature calculation device according to Item 1. The face orientation determination unit
The direction of the occupant's face is determined to be at least one of a front direction, an oblique direction, and a horizontal direction based on a center position of the occupant's face in the left-right direction within the face region. Item 3. The face temperature calculation device according to item 1 or 2. When it is determined that the direction of the occupant's face is a frontal direction,
4. The face temperature according to claim 3, wherein the contribution of the temperature of the peripheral region on one side in the left-right direction of the face region across the central region and the temperature of the peripheral region on the other side are made equal. Calculation device. When it is determined that the direction of the occupant's face is oblique, the contribution adjustment unit
The degree of contribution of the temperature of the peripheral area with the smaller area in the face area, out of the peripheral area on one side in the left-right direction of the face area across the central area and the peripheral area on the other side. 5. The face temperature calculation device according to claim 3, wherein the face temperature calculation device is set high. The contribution adjustment unit determines that the direction of the occupant's face is horizontal and the time determined to be horizontal is a predetermined time or more,
Of the peripheral region on one side in the left-right direction of the face region across the central region and the peripheral region on the other side, the degree of contribution of the temperature of the peripheral region with the larger area in the face region The face temperature calculation device according to claim 3, wherein the face temperature calculation device is set high. When the face temperature calculation unit determines that the direction of the occupant's face is horizontal and the time determined to be horizontal is less than a predetermined time,
The face temperature calculated immediately before it is determined that the direction of the occupant's face is sideways is defined as the occupant's face temperature. Face temperature calculation device. Generate a thermal image showing the temperature distribution within the detection range of the sensor, extract the occupant's face area from the generated thermal image, determine the orientation of the occupant's face in the face area, and determine the orientation of the determined face Accordingly, the face temperature of the occupant is calculated based on the temperature distribution in the face area after adjusting the contribution of the temperature of the central area of the face area and the temperature of the peripheral area. Calculation method.

Images