JP2016048128A - Air conditioner - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve the comfort of a person in a room on the basis of a position of his/her foot.SOLUTION: An air conditioner for sending cool air or warm air includes: an image acquisition part which takes an image of an area on the periphery of the air conditioner; a detection part which detects a person's head from the taken image, and which detects his/her foot within a predetermined range from his/her head; and a control part which adjusts the air to be sent, on the basis of a position of his/her foot detected by the detection part.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an air conditioner.

  A conventional air conditioner includes a sensor that acquires room spatial information, a human body detection determination unit that determines a range in which a human body exists from the spatial information acquired by the sensor, and a range in which a human body exists that is determined by the human body detection determination unit The human body state determination unit that determines the state of the human body by analyzing the spatial information of the human body, and an airflow control unit that controls the airflow based on the state of the human body determined by the human body state determination unit, and the determined human body state That is, air flow control or temperature adjustment suitable for the position of the human head or feet or the posture of the human body is performed.

JP 2012-0421131 A

  In the prior art, there has been no air conditioner that detects the position of a person's foot and harmonizes air toward the detected foot position.

  Patent Document 1 describes that a person is detected from the temperature distribution and the standing position and the sitting position are determined based on the detected aspect ratio. However, since the foot and the head are detected independently, the furniture leg and the human foot cannot be distinguished from each other and erroneously detected.

  The present invention provides an air conditioner that improves the comfort of a person in a room by accurately detecting the position of a person's foot and adjusting the air sent toward the detected position of the person's foot. Objective.

  In order to solve the above-described problems, the present invention provides an air conditioner that sends cold air or warm air, an image acquisition unit that captures an image around the air conditioner, and a human head from the captured image. A detection unit that detects a person's foot within a predetermined range from the person's head, and a control unit that adjusts air to be sent based on the position of the person's foot detected by the detection unit; It has the air conditioner characterized by having.

  ADVANTAGE OF THE INVENTION According to this invention, the air conditioner which improves the comfort of the person who exists indoors can be provided by detecting the position of a person's leg | foot correctly.

  Problems, configurations, and effects other than those described above will be clarified by the following description of embodiments.

It is explanatory drawing which shows the structure of the air conditioner of a present Example. It is explanatory drawing which shows the cross section of the indoor unit of a present Example. It is a block diagram which shows the structure of the air conditioner of a present Example. It is a block diagram which shows the structure of the sensor part of a present Example. It is a flowchart which shows the process which detects the position of the leg | foot of a present Example. It is explanatory drawing which shows the method of detecting the leg | foot by the background difference method of a present Example. It is a flowchart which shows the method of detecting the leg | foot by the lump detection method A of a present Example. It is explanatory drawing which shows the outline tracking method of a present Example. It is a flowchart which shows the method of detecting the leg | foot by the lump detection method B of a present Example. It is explanatory drawing of the process which specifies the candidate of the area | region where the leg | foot is represented of a present Example. It is explanatory drawing which shows the template of a present Example.

  Hereinafter, examples will be described with reference to the drawings.

  FIG. 1 is an explanatory diagram illustrating a configuration of an air conditioner 10 according to the present embodiment.

  The air conditioner 10 is a device that harmonizes indoor air, such as cooling or heating, using, for example, heat pump technology. The air conditioner 10 includes an indoor unit 20, an outdoor unit 30, and a remote controller 11.

  The indoor unit 20 is installed in a room, a hallway, or the like, and harmonizes the air in the room by adjusting the air sent to the room. The indoor unit 20 is installed on a wall, ceiling, or floor.

  In addition, the indoor unit 20 of a present Example may be installed in the outdoors which are not enclosed by the wall and the ceiling, and may adjust the air ventilated only to a specific range. In the following, an example in which the indoor unit 20 is mainly installed indoors surrounded by a wall and a ceiling is shown, but the indoor unit 20 performs the same processing even when installed indoors.

  The outdoor unit 30 circulates heat with the indoor unit 20 by constructing a refrigerant cycle with the indoor unit 20. The outdoor unit 30 includes a fan 18 and a compressor 19 (described later) in order to construct a refrigerant cycle. The outdoor unit 30 is installed outdoors or outdoors. The indoor unit 20 and the outdoor unit 30 are connected by a refrigerant pipe and a communication cable.

  The remote controller 11 receives the setting value input by the user and sends the received setting value to the indoor unit 20. Accordingly, the user operates the air conditioner 10 using the remote controller 11. The remote controller 11 communicates with the indoor unit 20 by infrared rays, radio waves, communication lines, or the like.

  The indoor unit 20 includes an imaging unit 26, a temperature detection unit 27, a sensor unit 300, and an electrical component 400. The imaging unit 26 is a device that has a camera or the like that captures a range in which the indoor unit 20 sends wind, and acquires an image. The temperature detection unit 27 is an apparatus for acquiring images representing the surface temperature of objects such as indoor furniture, walls, and floors, and objects installed in a range where the indoor unit 20 sends air. The indoor unit 20 may include a device that measures room temperature, humidity, and the like in addition to the imaging unit 26 and the temperature detection unit 27.

  The sensor unit 300 is a device that analyzes data acquired from a measurement device included in the indoor unit 20. The electrical component 400 is a device that controls a method of harmonizing air.

  FIG. 2 is an explanatory diagram showing a cross section of the indoor unit 20 of the present embodiment.

  The indoor unit 20 includes a heat exchanger 13, a blower fan 12, left and right louvers (left and right wind direction plates; described later in FIG. 3) 17, upper and lower louvers (up and down wind direction plates) 14, front panel 15, electrical component 400, imaging unit 26, temperature The detection unit 27 and various measurement units are accommodated in the housing base 16. The housing base 16 is a case that houses a device that implements the functions of the indoor unit 20.

  The heat exchanger 13 has a plurality of heat transfer tubes. The heat transfer tube of the heat exchanger 13 exchanges heat between the air collected in the indoor unit 20 from the room by the blower fan 12 and the refrigerant flowing through the heat transfer tube. Thereby, the heat exchanger 13 is configured to cool or heat the air after heat exchange. The heat transfer tube of the heat exchanger 13 is connected to the outdoor unit 30 via the refrigerant pipe and constitutes a part of the refrigerant cycle.

  The blower fan 12 collects air and sends out air. Moreover, the ventilation fan 12 adjusts the wind speed of the air sent out.

  The left and right louvers 17 move so as to swing in the horizontal direction in order to control the horizontal direction of sending out air. This is because the base end side of the left and right louvers 17 is continuously rotated in the forward and reverse directions by the left and right louver motors with a rotating shaft vertically installed below or above the indoor unit 20 as a fulcrum. And as a result of installing so that the front end side of the left and right louvers 17 may face the front panel 15, the front end side of the left and right louvers 17 moves so as to swing in the horizontal direction.

  The upper and lower louvers 14 move so as to swing in the vertical direction in order to control the vertical direction in which air is sent out. This is because the base end side of the upper and lower louvers 14 is continuously rotated in the forward and reverse directions by the upper and lower louver motors with the rotation shafts provided at both ends in the longitudinal direction of the indoor unit 20 as fulcrums. Thereby, the front end side of the vertical louver 14 moves so as to swing in the vertical direction.

  The front panel 15 is installed so as to cover the front surface of the indoor unit 20. Here, the front surface of the indoor unit 20 is a side surface opposite to the wall side on which the indoor unit 20 is installed. The front panel 15 is rotated forward and backward by the front panel motor with the rotation axis at the lower end as a fulcrum. The front panel 15 may be fixed to the lower end of the indoor unit 20 without rotating.

  The indoor unit 20 collects indoor air in the indoor unit 20 through the air inlet and the filter as the blower fan 12 rotates. Then, the heat exchanger 13 exchanges heat with the collected air. The heat exchange is to absorb heat from the air or to release heat to the air. The air after heat exchange is guided to the blowout air path.

  Furthermore, the air guided to the blowout air passage is sent out of the indoor unit 20 through the air outlet. Thereby, the indoor unit 20 adjusts indoor air. When the air after heat exchange blows into the room from the outlet, the left and right louvers 17 adjust the horizontal wind direction, and the upper and lower louvers 14 adjust the vertical wind direction.

  The imaging unit 26 images the periphery of the air conditioner 10, specifically, the room where the indoor unit 20 is installed. In addition, when the indoor unit 20 is installed above the room, the imaging unit 26 photographs the front and lower sides of the indoor unit 20, and when the indoor unit 20 is installed below the indoors, An apparatus for photographing the upper part, for example, having a CCD (Charge Coupled Device) image sensor. The imaging unit 26 may be installed at any position of the indoor unit 20 as long as the room can be imaged. The imaging unit 26 in the present embodiment is installed in the lower part of the center in the longitudinal direction of the front panel 15.

  Similar to the imaging unit 26, the temperature detection unit 27 is a device that acquires the surface temperature around the air conditioner 10. For example, the horizontal x vertical is 1 x 1 pixel, 4 x 4 pixels, or 1 x 8 It is a thermopile composed of pixels. The temperature detection unit 27 may be installed at any position of the indoor unit 20 as long as the indoor surface temperature can be acquired. The temperature detection unit 27 in the present embodiment is installed at the lower part of the center of the front panel 15 in the longitudinal direction, and is arranged next to the imaging unit 26.

  In the following description, the imaging unit 26 has a CCD image sensor that measures a visible light image, and the temperature detection unit 27 has a thermopile for measuring the surface temperature. However, the imaging unit 26 and the temperature detection unit 27 may be any device as long as it measures the physical quantity of the surface of an object in the room and generates data that can display a change in the measured value in an image that can be visually recognized by humans. There may be.

  For example, the imaging unit 26 and the temperature detection unit 27 may be an infrared sensor, a near-infrared sensor (a near-infrared irradiation LED and a sensor capable of photographing a light environment including the near-infrared), a thermography, a pyroelectric sensor, a noise sensor, or a super You may have a sonic sensor.

  The temperature detection unit 27 acquires the average surface temperature of the room, the surface temperature of the person, and the surface temperature of the room in the area excluding the person. For example, the temperature detection unit 27 acquires a human clothing and skin surface temperature, and a floor surface temperature.

  The indoor unit 20 may include a sensor that measures room temperature and a sensor that measures humidity in addition to the imaging unit 26 and the temperature detection unit 27.

  FIG. 3 is a block diagram illustrating a configuration of the air conditioner 10 according to the present embodiment.

  In the indoor unit 20, the sensor unit 300 is connected to the imaging unit 26, the temperature detection unit 27, and a measurement device such as a device that measures room temperature or room humidity, and is acquired by the imaging unit 26 and the temperature detection unit 27. Receive data. The sensor unit 300 detects the position of the head of a person in the room based on the received data, and further detects the position of the foot based on the detected position of the head of the person.

  The sensor unit 300 notifies the detected position of the foot to the control unit 330 of the electrical component 400. The control unit 330 determines the wind direction and the wind speed based on the notified contents. And according to the determined wind direction and wind speed, operation | movement is instruct | indicated to the apparatus provided in the indoor units 20, such as the ventilation fan 12, the left-right louver 17, and the upper-lower louver 14, and the fan 18 and the compressor 19 which are provided in the outdoor unit 30.

  The devices such as the blower fan 12, the left and right louvers 17 and the upper and lower louvers 14, and the devices such as the fan 18 and the compressor 19 harmonize air according to instructions from the control unit 330 and send the conditioned air into the room.

  The sensor unit 300 and the electrical component 400 may be two independent physical devices each having a processor and a memory, and the sensor unit 300 and the electrical component 400 may be mounted on one device having a processor and a memory. Good.

  The compressor 19 is a device that compresses the refrigerant. The fan 18 is a device that circulates outdoor air in order to dissipate heat or absorb heat from the outdoor.

  In addition, the air conditioner 10 includes devices such as an expansion valve that depressurizes the high-pressure refrigerant and a four-way valve that switches the flow path of the refrigerant. However, since a general technique is used in these embodiments for these configurations, description thereof is omitted.

  FIG. 4 is a block diagram illustrating a configuration of the sensor unit 300 according to the present embodiment.

  The sensor unit 300 includes a processor 310, a memory 311, and an interface 320. The processor 310 is an arithmetic device and a control device, for example, a CPU. The interface 320 is an interface for transmitting and receiving signals to and from the measurement device provided in the electrical component 400 or the indoor unit 20.

  The memory 311 is a storage area that holds data. The processor 310 implements the function of the sensor unit 300 by developing a program in the memory 311.

  The memory 311 holds a human detection unit 301, a foot detection unit 302, a temperature acquisition unit 303, a room temperature acquisition unit 304, a humidity acquisition unit 305, and a timer 306 as programs. The plurality of programs stored in the memory 311 may be implemented as a single program, or may be divided for each process included in each program.

  The person detection unit 301 detects a person in the room based on the data transmitted from the imaging unit 26 and the temperature detection unit 27, and further detects the position of the person's head.

  The foot detection unit 302 detects the presence / absence of a human foot, the position of the foot, and the number of feet based on the result detected by the human detection unit 301. In addition, the foot detection unit 302 may output the likelihood of the detection result. The temperature acquisition unit 303 acquires the indoor surface temperature based on the image.

  The room temperature acquisition unit 304 acquires the room temperature from a device that measures the room temperature installed in the indoor unit 20. The humidity acquisition unit 305 acquires the indoor humidity from a device that measures the indoor humidity installed in the indoor unit 20.

  The timer 306 holds time. When the sensor unit 300 receives an image sent from the imaging unit 26 and the temperature detection unit 27, the received time is held. Note that the timer 306 may add the received time information to the image held in the memory 311.

  Details of the functions of the imaging unit 26 and the temperature detection unit 27 will be described below.

  The imaging unit 26 and the temperature detection unit 27 of the present embodiment measure the same range in the room. Further, the upper side and the lower side of the imaging results obtained by the imaging unit 26 and the temperature detection unit 27 correspond to the ceiling-side wall and floor in the room, respectively. Each processing unit of the sensor unit 300 can specify the position and temperature of the object in the room by superimposing the image acquired by the imaging unit 26 and the image acquired by the temperature detection unit 27.

  When the indoor unit 20 is installed on the upper side of the room, the imaging unit 26 and the temperature detecting unit 27 are installed so that the optical axes of the lenses included in the indoor unit 20 face downward within a predetermined angle range with respect to the horizontal line. Is done. This is because the imaging unit 26 and the temperature detection unit 27 can appropriately measure the room.

  The angles at which the imaging unit 26 and the temperature detection unit 27 face downward are substantially the same. In addition, when the vertical range that can be measured by the imaging unit 26 and the vertical range that can be measured by the temperature detection unit 27 are different, for example, the upper end of the range that the imaging unit 26 measures and the range that the temperature detection unit 27 measures The imaging unit 26 and the temperature detection unit 27 are installed so that the upper ends of the two coincide with each other.

  This is because the ranges measured by the imaging unit 26 and the temperature detection unit 27 are matched. In addition, you may make the lower end of the range to measure match.

  Further, the horizontal field angles of the imaging unit 26 and the temperature detection unit 27 are substantially the same. Further, when one is larger than the other, a measurement unit having a small angle of view may rotate and measure, so that a range of almost the same angle of view as a measurement unit having a large angle of view may be measured.

  The imaging unit 26 includes, for example, a 640 × 480 pixel CCD image sensor, and the temperature detection unit 27 includes, for example, a 1 × 8 pixel thermopile. A lens is provided in front of the CCD image sensor and the thermopile, and a field image is formed on the sensor.

  The CCD image sensor is a two-dimensional image sensor. In order to widen the image range that can be measured by the imaging unit 26, the CCD image sensor is rotated about the vertical direction of the CCD image sensor as a rotation axis, and the horizontal direction is scanned. As a result, when the horizontal range that the CCD image sensor can capture at one time is not sufficient, the imaging unit 26 obtains an image larger than the number of pixels in the horizontal direction of the CCD image sensor.

  The detection element of the thermopile is a heat receiving element arranged one-dimensionally. By rotating the thermopile with the arrangement direction of the detection elements as a rotation axis, scanning is performed in a direction perpendicular to the arrangement direction of the detection elements. Thereby, the temperature detection unit 27 can acquire a two-dimensional radiant heat image of eight pixels in the vertical direction.

  For example, the CCD image sensor of the imaging unit 26 has a field angle of 60 °, and the thermopile has a field angle of 5 °. And the imaging part 26 and the temperature detection part 27 of a present Example acquire the image of the same visual field with an angle of view whose horizontal direction is 150 degrees.

  Since the CCD image sensor of the imaging unit 26 acquires an image having a field angle of 150 ° in the horizontal direction, one image representing the room is obtained by imaging at three rotation angles of 45 °, left, center, and right. To get. Since the thermopile of the temperature detection unit 27 acquires an image having a field angle of 150 °, the thermopile acquires a radiant heat image in a range of 75 ° at each of the center, the left, and the right at every rotation angle of 5 °. .

  The imaging unit 26 acquires an image having a field angle of 150 ° in the horizontal direction and a pixel number of 1600 × 480. The temperature detection unit 27 acquires a thermal image having a field angle of 150 ° in the horizontal direction and a pixel number of 30 × 8. At this time, since there are overlapping areas in the plurality of images captured by the image capturing unit 26, the image capturing unit 26 appropriately deletes or averages and acquires the image having the number of pixels of 150 ° in the horizontal direction described above.

  Note that the number of pixels of the acquired image is an example, and the number of pixels varies depending on the angle of view to be acquired and the type of CCD image sensor or thermopile to be used.

  The imaging unit 26 and the temperature detection unit 27 are arranged adjacent to each other in the horizontal direction or the vertical direction. This is because when the base point to be measured is greatly different, the range and orientation to be measured are different, and the difference between the image acquired by the imaging unit 26 and the image acquired by the temperature detection unit 27 is reduced.

  Next, a mechanism for rotating the thermopile or the CCD image sensor will be described. The driving source for rotating the thermopile or the CCD image sensor is a stepping motor in this embodiment.

  A rotating shaft that rotates the thermopile or the CCD image sensor and a drive shaft of the stepping motor are connected via a gear. Even for a stepping motor with a small number of steps, the rotation angle as described above is set by appropriately setting the gear ratio.

  A rotary shaft that rotates the thermopile or the CCD image sensor and a drive shaft of the stepping motor may be connected by a four-bar link to convert the rotary motion into a swing motion. When the drive shaft is connected by the above-described gear or link, an error may occur in the rotational position accuracy due to gear backlash or link “play”.

  For this reason, in order to prevent an error of the rotation angle due to “play”, the manufacturer may set the rotation direction at the time of imaging of the thermopile or the CCD image sensor to one direction. In addition, the manufacturer may obtain an adjustment amount for absorbing “play” in advance and set the adjustment amount when the rotation direction is reversed.

  FIG. 5 is a flowchart illustrating processing for detecting the position of the foot according to the present embodiment.

  The imaging unit 26 and the temperature detection unit 27 capture the room (501). The imaging unit 26 and the temperature detection unit 27 send one image having a field angle of 150 ° in the horizontal direction acquired by photographing to the sensor unit 300 as image data. In the following, an image acquired by the imaging unit 26 is particularly referred to as a visible light image, and an image acquired by the temperature detection unit 27 is particularly referred to as a temperature image.

  After step 501, the person detection unit 301 detects the position of the person's head in the room based on the received image (502). The human detection unit 301 may detect the position of the human head using any method.

  Note that the human detection unit 301 and the foot detection unit 302 in the following processing may use a visible light image or a temperature image in order to detect a person and a foot. However, the images used by the human detection unit 301 and the foot detection unit 302 are images indicating physical quantities such as luminance or temperature on the surface of a person, furniture, wall, door, or the like. In the following, processing when the visible light image acquired by the imaging unit 26 is used will be mainly described.

  For example, the person detection unit 301 may detect the position of a circle included in the room as the position of the person's head (that is, the position of the person) assuming that the person's head is a circle. In addition, the human detection unit 301 holds a human face eye, nose, and mouth pattern in advance, and when a portion corresponding to the facial pattern is detected, the human detection unit 301 determines the position of the human facial pattern as the human head. It may be detected as a position (that is, a person's position).

  If no person is represented in the visible light image, the person detection unit 301 waits for the next reception of the visible light image.

  The human detection unit 301 compares the previously received visible light image (hereinafter, image 2) with the currently received visible light image (hereinafter, image 1), and calculates the amount of movement by which the detected person has moved ( 503). The memory 311 holds at least one visible light image received in the past. The human detection unit 301 acquires a visible light image received last time or a visible light image received in the past for a predetermined time from the memory 311 as an image 2 as a visible light image for comparison with the image 1.

  In step 503, the human detection unit 301 calculates the distance between the position of the human head detected from the image 2 and the position of the human head detected from the image 1, and uses the calculated distance as the amount of movement. get. Further, when no person is represented in the image 2, the person detection unit 301 may determine a value larger than a threshold value (described later) used in step 504 as the amount of movement.

  After step 503, the human detection unit 301 determines whether or not the calculated amount of motion is greater than a predetermined distance threshold (504). The threshold value used in step 504 is a distance that a person may move in the time between the time when the image 2 is acquired and the time when the image 1 is acquired. That is, the person detection unit 301 holds in advance a relationship between the time between the time when the image 2 is acquired and the time when the image 1 is acquired and the distance that the person may move.

  This is based on the assumption that an object that moves at a constant speed is a person or an object that moves by the movement of a person. Furthermore, non-living inanimate objects move at a speed at which a person walks or a person runs. Based on the assumption that there is little to do.

  When the calculated amount of motion is greater than a predetermined distance threshold, the person detection unit 301 determines that the person represented by the image 1 is moving, and uses a background difference method and two images described later. And decide to detect the foot. The foot detection unit 302 detects the foot using a method according to the determination of the human detection unit 301 (506).

  When the calculated amount of motion is smaller than a predetermined distance threshold, the human detection unit 301 determines to detect a foot using a mass detection method and a single image, which will be described later. The foot detection unit 302 detects the foot using a method according to the determination of the human detection unit 301 (505).

  The background difference method tends to require less processing time than the lump detection method. In addition, since a moving object is likely to be a living thing, and a living thing in a room is likely to be a human, the background subtraction method can detect a person with higher accuracy than the lump detection method.

  For this reason, the human detection unit 301 executes steps 503 and 504 to determine a method for detecting a person and a foot so that the processing time is faster and more accurate than when only the lump detection method is used. A method for detecting a person and a foot can be selected.

  After step 505 or step 506, the foot detection unit 302 notifies the control unit 330 of the electrical component 400 of the detected foot position. And the control part 330 adjusts the air sent to the position of a foot based on the notified position of the foot (507).

  In addition to the position of the foot, the foot detection unit 302 may notify the control unit 330 of the number of detected feet and the likelihood of the detection result. Further, the temperature acquisition unit 303 may notify the control unit 330 of the surface temperature of the foot position, the surface temperature around the foot position, and the like according to the result and instruction of the foot detection unit 302.

  Details of the adjustment of air flow by the controller 330 will be described later.

  In the process shown in FIG. 5, the air conditioner 10 according to the present embodiment detects the foot based on the position of the person's head, and thus can accurately detect the person's foot. And the comfort of the person who is indoors can be raised by adjusting the air sent to the position of the detected person's foot.

  FIG. 6 is an explanatory diagram showing a method of detecting a foot by the background difference method of the present embodiment.

  The background subtraction method is a method for detecting the position of the foot based on the position of the moving object assuming that the moving object is a person.

  The foot detection unit 302 acquires the image 1 and the image 2 from the memory 311. Then, the foot detection unit 302 detects a human head from each of the image 1 and the image 2 (601).

  The method for detecting the human head is the same as the method for detecting the human head used in step 502. Note that the foot detection unit 302 may not detect the person's head from the image 2 in step 601 when the person is not represented in the image 2.

  After step 601, the foot detection unit 302 performs a smoothing process on the contents shown in the image 1 and the image 2 (602). Here, the smoothing process refers to a process of overwriting noise represented by a minute area not having an area of a predetermined size with a surrounding color (physical quantity: luminance, temperature, or the like) in the image. . For example, when a plurality of small dots of a color different from most of the wall positions are displayed at a certain position, the foot detection unit 302 overwrites the dot portion with the same color as the surrounding wall of the point. To do.

  By the smoothing process, the foot detection unit 302 can accurately detect the contour lines (including straight lines or curves) of the objects (furniture, people, etc.) represented in the images 1 and 2.

  After step 602, the foot detection unit 302 extracts a human region represented in the image 1 by extracting the difference between the contents represented in the image 1 and the image 2 (603). Here, the content represented in the image is a physical quantity represented in the image, and is an object (surface has a physical quantity) represented in the image.

  In step 603, the foot detection unit 302 specifically compares the image 1 and the image 2, and extracts a region having contents not represented in the image 2 from the image 1. Then, the foot detection unit 302 extracts a region including the position of the head detected from the image 1 in step 601 among the extracted regions as a region representing a person.

  After step 603, the foot detection unit 302 detects the position of the foot in the region where the person is represented using the foot estimation filter (604). Here, the foot estimation filter is a function that estimates the position of the foot from an area where a person is represented using at least one of the following estimation methods 1 to 3.

  The estimation method 1 is a method for estimating a lowermost predetermined range (determined by at least one of a predetermined area, height, and width) in a region where a person is represented as a region representing a foot.

  The estimation method 2 is a method of estimating, in a region where a person is represented, a region separated by a predetermined distance or area from the position where the head is detected as a region representing a foot.

  In the estimation method 3, the lowermost region (which may be a region with a predetermined area) of the region that is separated from the position where the head is detected by a predetermined distance or area or more in the region where the person is represented is This is a method of estimating the represented area.

  In step 604, the foot detection unit 302 estimates a region representing the foot using at least one of the estimation methods 1 to 3 described above, and acquires the position of the estimated region as the foot position. The foot detection unit 302 can accurately detect a human foot by using at least one estimation method described above.

  The foot detection unit 302 uses the predetermined range used in the estimation method 1 and the predetermined distance and area used in the estimation method 2 and the estimation method 3 as the distance to the person detected in step 603, step 603. Frequency at which a person was detected in the past at the position where the person was detected, the amount of movement calculated in step 503, the number of detected persons, room temperature, wall and floor surface temperatures, illuminance, luminance, and time zone Based on the above, it may be determined before step 604.

  In this case, the person detection unit 301 or the like may obtain the distance from the indoor unit 20 to the person based on the area of the person's head in the image 1. For example, the human detection unit 301 holds in advance a combination of the area on the head image when the head having a standard human cross-sectional area is represented in the image and the distance to the indoor unit 20. May be. Then, the human detection unit 301 may obtain the distance according to the held combination with the area of the human head in the image 1.

  Further, the room temperature acquisition unit 304 may acquire the room temperature. Furthermore, the indoor unit 20 may include a measurement unit that measures the illuminance and luminance in the room.

  For example, the foot detection unit 302 may determine the predetermined range, distance, and area based on the distance to the detected person and the average adult height, foot size, and body cross-sectional size. May be calculated.

  Further, when using the background difference method, the foot detection unit 302 may determine the predetermined distance based on the amount of motion calculated in step 503. Specifically, since the foot detection unit 302 is more likely to stand if the amount of motion is large, the predetermined distance in the estimation methods 2 and 3 is set large, and if the amount of motion is small or zero, the foot detection unit 302 estimates The predetermined distance in the methods 2 and 3 may be set small.

  Further, the foot detection unit 302 finally detects and outputs a region acquired as a region representing a foot by a plurality of estimation methods among the estimation methods 1 and 2 as a foot position. Also good. In addition, the foot detection unit 302 may calculate the probability of the detection result based on the number of estimation methods used. Then, the foot detection unit 302 may notify the control unit 330 of the calculated certainty factor, and the control unit 330 may adjust the air sent out based on the certainty factor.

  After step 604, the foot detection unit 302 outputs a detection result (including the number of feet and the probability of the detection result) such as the detected foot position, and ends the processing shown in FIG. By the background subtraction method shown in FIG. 6, the foot detection unit 302 accurately detects the foot by detecting the foot using two images based on the assumption that the moving object is likely to be a person. can do.

  Hereinafter, the lump detection method executed in step 505 will be described. There are two lump detection methods, lump detection method A and lump detection method B. The foot detection unit 302 detects a foot using either one of the lump detection method A and the lump detection method B. The lump detection method A and the lump detection method B are each a method of acquiring the position of the foot from one visible light image.

  FIG. 7 is a flowchart showing a method of detecting a foot by the lump detection method A of the present embodiment.

  The foot detection unit 302 first acquires the image 1 from the memory 311. The foot detection unit 302 detects the human head from the image 1 (611). The method of detecting the human head in step 611 is the same as the method used in step 601 of FIG.

  After step 611, the foot detection unit 302 performs a smoothing process on the image 1 (612). The smoothing process in step 612 is the same as the smoothing process in step 611 of FIG.

  After step 612, the foot detection unit 302 detects an edge in the image 1 (613). Specifically, the foot detection unit 302 detects an area between areas in which the luminance of the image 1 greatly changes as an edge. When the image 1 is a temperature image, an area between regions in which a physical quantity such as the temperature of the image 1 changes greatly (specifically, a difference larger than a predetermined threshold value) may be detected as an edge.

  After step 613, the foot detection unit 302 divides the area included in the image 1 using the detected edge (614). This is a process for dividing an image for each object represented in the image 1.

  The foot detection unit 302 may use any method as long as the region can be divided in step 614. For example, the foot detection unit 302 may divide the region using a contour tracking method.

  FIG. 8 is an explanatory diagram showing the contour tracking method of the present embodiment.

  The contour tracking method is a method of assuming that an area surrounded by the extracted edges is an area representing an object when edges are sequentially extracted from the start point on the edge and returned to the start point.

  A region 701 indicated by hatching in FIG. 8 is surrounded by an edge detected from the image 1. Note that the image 1 is divided into blocks having a predetermined size.

  The foot detection unit 302 raster scans the image 1 and detects one block included in the region 701. The foot detection unit 302 extracts the first detected block as the start point, and extracts the edge end block from the start point block counterclockwise. Here, the foot detection unit 302 searches for an end in the order of lower left, lower, lower right, right, and upper right, and extracts an end block of the region 701. When returning to the start point block again, the foot detection unit 302 detects the extracted block as the contour of the region 701.

  Thereafter, the foot detection unit 302 extracts a block that has not been extracted as a start point as a start point, and detects a contour by repeating the above-described processing.

  By extracting the block at the end of the region 701 in this way, the foot detection unit 302 detects the contour 702 and the contour 703. Then, when there is no edge between the contour 702 and the contour 703, the foot detection unit 302 divides the inside of the contour 702 and the contour 703 from other regions as a region where an object is represented. As a result of the division, the foot detection unit 302 acquires a plurality of regions.

  After step 614, the foot detection unit 302 calculates feature amounts indicating the features of the plurality of divided regions, respectively (615). The feature amount in this embodiment is an index that quantitatively indicates the feature of the region, and is, for example, the circularity, area, or complexity of the region.

  Circularity is an index indicating whether or not a region is close to a circle (circularness). The circularity is a numerical value from 0 to 1, for example, and the closer to 1, the closer to a circle. The circularity is generally calculated using the area and the perimeter.

  In addition, the foot detection unit 302 calculates the complexity based on a change in a physical quantity displayed by the image, such as the number of edges included in the divided area. The complexity in this embodiment is, for example, the density of edges included in a region, and is the amount of change such as temperature or luminance represented by an image. The complexity of the present embodiment indicates that the more the number of edges included in the region is, the higher the amount of change is, the more complicated it is.

  After step 615, the foot detection unit 302 removes noise from the divided regions based on at least one of the calculated feature values (616). The noise in step 616 is an area representing an object (furniture, floor, wall, etc.) that is not a person.

  For example, the foot detection unit 302 extracts a circularity region between a predetermined maximum and minimum circularity as a region where a person is represented. This is because a person is more likely to be round than square furniture and the like, and is more likely to be less round than a perfect circle window.

  In addition, for example, the foot detection unit 302 extracts a region having an area between a predetermined maximum value and a minimum value as a region where a person is represented. This is because the size of a person is likely to be statistically included in a certain range, and the size of a person photographed by the imaging unit 26 is also likely to be included in a certain range.

  In addition, for example, the foot detection unit 302 extracts a region having a calculated complexity higher than a predetermined threshold as a region where a person is represented. This is because the surface of the person is bumpy and can be highly complex.

  Then, the foot detection unit 302 removes, as noise, a region that has not been extracted as a region where a person is represented as described above. Note that the foot detection unit 302 may extract a region where a person is represented using the plurality of feature amounts described above.

  By extracting the region in which the person is represented based on the feature amount of the region, the foot detection unit 302 can narrow down the region in which the person is represented, and can detect the foot more accurately.

  After step 616, the foot detection unit 302 obtains the position of the foot in the area where the person is represented using the foot estimation filter (617). The processing in step 617 is the same processing as step 604 in FIG. After step 617, the foot detection unit 302 outputs a detection result (including the number of feet and the likelihood of the detection result) such as the position of the detected foot, and ends the process shown in FIG.

  FIG. 9 is a flowchart showing a method of detecting a foot by the lump detection method B of the present embodiment.

  Steps 621 to 623 are the same as steps 611 to 613 shown in FIG.

  After step 623, the foot detection unit 302 binarizes the image 1 based on the detected edge strength (624). In step 623, the foot detection unit 302 detects an edge having a stronger intensity as the luminance change amount between the regions is larger. Then, as a binarization process, the foot detection unit 302 deletes an edge having a strength lower than a predetermined threshold in Step 624.

  Accordingly, the foot detection unit 302 can delete edges other than the edges due to the contour of the object, such as a pattern of the object or a small distortion of the surface of the object. The foot detection unit 302 can divide the image 1 into an object region and a background region.

  After step 625, the foot detection unit 302 specifies an outline surrounding the position of the head detected in step 621 in the image 1. Then, the foot detection unit 302 searches for the periphery of the region surrounded by the specified contour using a template described later, and specifies a candidate region where the foot is represented.

  FIG. 10 is an explanatory diagram of a process for narrowing down candidates for the area in which the foot is represented according to this embodiment.

  First, the foot detection unit 302 in step 625 specifies an area 801 in which the human head is represented in the image 1. Then, the foot detection unit 302 blocks the image 1 based on the luminance. Specifically, the image 1 is divided into a plurality of blocks, the luminance represented in the block is statistically processed, and the luminance of the block is determined. Then, the foot detection unit 302 generates a block image by assigning the determined luminance of the block to each block.

  An image 800a illustrated in FIG. 10 is an image obtained by dividing the image 1 into a plurality of images. An image 800b illustrated in FIG. 10 is a block image generated by determining the luminance based on the luminance of the image 800a. The block of the image 800b corresponds to the block of the image 800a.

  In addition, the foot detection unit 302 may use any method as the statistical processing described above. For example, as a statistical process, the foot detection unit 302 may calculate an average value of physical quantities represented by one block of the image 800a and determine the calculated average value as a luminance assigned to the block.

  In addition, when one block of the image 800a represents a physical quantity that is greater than or equal to a predetermined threshold value by 30% or more, the foot detection unit 302 may determine the luminance assigned to that block as the physical quantity that is used as the predetermined threshold value. . Then, when the luminance equal to or higher than the predetermined threshold is less than 30%, the foot detection unit 302 may assign the minimum physical quantity value represented by the block to the physical quantity of the block in the block image.

  By making the image 1 into a block, the foot detection unit 302 can simplify the shape of the region where the person is represented, and easily use the template having a simple shape such as a template described later to easily represent the region where the foot is represented. Can be extracted.

  FIG. 11 is an explanatory diagram showing a foot-shaped template according to this embodiment.

  The foot detection unit 302 uses a template such as the template 802, the template 803, or the template 804 to identify a candidate for a region where the foot is represented. The template shown in FIG. 11 is a concave shape in which one side of a quadrangle is opened. The outer side 806 is the outer contour of the template. An inner side 805 is an inner contour of the template.

  In particular, a person's foot is likely to be an end in a direction in which a region where the person is represented extends. For this reason, the template shown in FIG. 11 is an example of a template suitable for extracting the end of the region.

  Note that the template of the present embodiment may be a template having any shape other than the concave shape as long as the template indicates the shape of a person's foot. For example, the template of the present embodiment may be a template that indicates a state where the legs are open.

  Further, the template according to the present embodiment may be inclined stepwise so that the contour of the inner side 805 is close to a curved line. The sizes of the outer side 806 and the inner side 805 of each of the template 802, the template 803, and the template 804 are determined in advance according to the size of the head region 801.

  The template 802 is used when a region where a person is represented is vertically long, and the templates 803 and 804 are used when a region where a person is represented is long horizontally.

  After generating the block image, the foot detection unit 302 extracts a region representing a person by extracting a region having a luminance close to the luminance of the region 801 in the image 1 from the region 801 of the head. Specifically, the foot detection unit 302 extracts a luminance region between a luminance obtained by adding a predetermined value to the luminance of the region 801 and a luminance obtained by subtracting a predetermined value from the luminance of the region 801. Extract regions where people are represented.

  An area corresponding to the template 802, the template 803, or the template 804 is searched. In addition, since the area | region where the person is represented in the image 800b is long vertically, the foot | leg detection part 302 may search using the template 802. FIG.

  Below the image 800b corresponds to the concave shape of the template 802, and a region having the same luminance as the luminance of the region where a person is represented is interrupted. Here, the inner side 805 of the template 802 is a foot portion of an area where a person is represented, and the outer side 806 is a background area. For this reason, the foot detection unit 302 identifies a region corresponding to the inner side 805 of the region corresponding to the template 802 as a candidate region where the foot is represented.

  In addition, when the region having the luminance closest to the luminance of the region representing the head continues laterally from the region 801 of the head, the foot detection unit 302 uses the template 803 or the template 804 to identify the region where the luminance is interrupted. To detect. This is because a person may be lying down.

  As described above, the foot detection unit 302 can narrow down the region representing the foot in step 625 by specifying the candidate region of the region representing the foot using the template. The foot detection unit 302 can detect the foot more accurately.

  After step 625, the foot detection unit 302 detects the position of the foot using the same foot estimation filter as in step 604 (626). However, the foot detection unit 302 in step 626 is a region included in the candidate region specified in step 625 and the region detected using each of the estimation methods 1 to 3 described above is detected by the foot. Detect as represented area.

  For example, the foot detection unit 302 uses the candidate region specified in step 625 and the estimation method 1 to detect the lowermost predetermined range in the candidate region as the region where the foot is represented.

  After step 626, the foot detection unit 302 outputs a detection result (including the number of feet and the probability of the detection result) such as the position of the detected foot, and ends the process shown in FIG. By using the lump detection method B, it is possible to detect a person and a foot from one image.

  Details of the processing of the control unit 330 will be described below.

  The control unit 330 controls the output of the air conditioner 10 so as to change the state of the foot to a target state that the person feels comfortable according to the notification from the foot detection unit 302. Specifically, for example, when the air conditioner 10 performs a heating function, the upper and lower louvers 14 and the left and right louvers 17 are configured to send warm air toward the foot position detected by the foot detection unit 302. Adjust at least one of the. As a result, the controller 330 can directly warm a person's foot. Moreover, the control part 330 can warm the floor near a leg | foot by ventilating toward the position of a leg | foot, and can improve the comfort of the person who touches there.

  Further, the control unit 330 may adjust at least one of the upper and lower louvers 14 and the left and right louvers 17 so as to send warm air toward a position just before the foot position detected by the foot detection unit 302. That is, the control unit 330 strikes the airflow sent from the indoor unit 20 against the floor in front of the feet without directly hitting the human body. This makes it possible to supply warm air at a reduced speed to a person in a state where the feeling of airflow (feeling of airflow) is reduced. And thereby, the control part 330 can improve a person's comfort.

  In addition, the control unit 330 can further enhance human comfort by changing the wind speed based on the position of the foot detected by the foot detection unit 302. Specifically, when the distance from the indoor unit 20 to the foot is longer than a predetermined threshold, the control unit 330 increases the rotation speed of the blower fan 12 and increases the wind speed, thereby reliably ensuring the airflow on the foot. May be delivered.

  Here, the controller 330 may calculate the distance from the indoor unit 20 to the foot by any method. The control unit 330 uses, for example, the ratio of the head width in the image 1 to the average human head diameter, and the length in the longitudinal direction of the region in which the person is represented in the image 1. You may calculate your height. Then, the control unit 330 uses the calculated height of the person, the distance from the indoor unit 20 to the person's head, the height at which the indoor unit 20 is installed, and the Pythagorean theorem, from the indoor unit 20 to the foot. The distance may be calculated.

  Further, when the position of the foot is not at the front of the indoor unit 20 but at the right side or the left side, the wind speed decreases due to the deflection of the air flow by the left and right louvers 17. For this reason, when the position of the foot is in a position biased to the right side or the left side of the indoor unit 20, the control unit 330 increases the rotation speed of the blower fan 12 and increases the wind speed, thereby reliably delivering the airflow to the foot. Also good.

  Furthermore, the temperature acquisition unit 303 may detect the surface temperature of the foot or the temperature around the foot based on the position of the foot detected by the foot detection unit 302. And according to the notification from the temperature acquisition part 303, the control part 330 may adjust ventilation.

  Here, the circumference | surroundings of a leg are a floor, a wall, etc., and are the environment which can change the state of a leg | foot. The floor supplies heat directly to the feet and takes heat away from the feet. In addition, the wall supplies heat to the foot or takes heat away from the foot through the air in the room.

  Specifically, based on the surface temperature of the foot and the temperature around the foot indicated by the notification from the temperature acquisition unit 303, the control unit 330 promotes heat radiation from the foot, It may be determined whether the environment is a cooling environment or the environment around the foot promotes heat radiation to the foot and warms the foot. Then, the control unit 330 determines at least one of the wind direction, the wind speed, the wind direction time, the blown air temperature, and the like according to whether the air conditioner 10 functions as heating or cooling and according to the above-described determination result. You may adjust.

  For example, when the notification from the temperature acquisition unit 303 indicates that the temperature around the foot is lower than a predetermined temperature, and the air conditioner 10 functions as heating, the control unit 330 determines that the environment around the foot is It is an environment that cools the foot, and it is determined that the foot is not in the target state, and air blowing is controlled. Specifically, the control unit 330 increases the set value of the blown air temperature (described later), turns the wind direction toward the foot, narrows the swing range, increases the wind speed, and lengthens the wind direction time for supplying the airflow to the foot. The environment around the foot may be changed to an environment that warms the foot. Thereby, the control part 330 can improve a person's comfort.

  Furthermore, the control unit 330 may adjust the ventilation according to the output of the human detection unit 301. Specifically, when a person is detected by the human detection unit 301 but no foot is detected by the foot detection unit 302, the control unit 330 estimates the position of the foot from the position of the person's head, and the estimated foot Based on the position, at least one of the wind direction, the wind speed, the wind direction time, the blown air temperature, and the like may be adjusted.

  In addition, when the human detection unit 301 or the foot detection unit 302 detects a plurality of people or a plurality of feet in the room, the control unit 330 moves the left and right louvers 17 to a region between the feet and the feet, or You may adjust so that it may carry out a swing driving | running | working so that the area | region between persons, or a several leg | foot or a person may be included by all means. In addition, the control unit 330 may adjust the upper and lower louvers 14 so as to blow air based on the feet of the closest person.

  Further, the control unit 330 may notify the likelihood of the detection result from the foot detection unit 302, and may reduce the degree of adjustment when the notified probability indicates that the detection result is not reliable. For example, when notified that the detection result is not certain, the controller 330 reduces the amount of change in the adjustment temperature of the air.

  The wind direction, wind speed, wind direction time, and blown air temperature in the present example are shown below.

  The wind direction in the present embodiment is controlled by at least one of the upper and lower louvers 14 and the left and right louvers 17. The wind direction is the direction of air sent out from the indoor unit 20. The controller 330 may fix the wind direction in only one direction using at least one of the upper and lower louvers 14 and the left and right louvers 17 or may perform a swing operation.

  Here, swing operation refers to continuously swinging the wind direction, temporarily stopping the swing at a predetermined timing, swinging the wind direction only at a predetermined timing, updating the range to swing at a predetermined timing, It refers to driving that swings based on the position of a person's head or foot, or that swings at a predetermined cycle such as 1 / f.

  The wind speed in the present embodiment indicates the speed of air sent from the indoor unit 20 by adjusting the rotation speed of the blower fan 12 and the angle and direction at which at least one of the upper and lower louvers 14 and the left and right louvers 17 is opened.

  The controller 330 uses at least one of the blower fan 12, the upper and lower louvers 14, and the left and right louvers 17 to fix or change the wind speed. Specifically, the control unit 330 adjusts the wind speed so as to fluctuate at a predetermined cycle such as 1 / f, updates the wind speed at a predetermined timing, or adjusts the position of the human head or foot. Based on this, the wind speed is adjusted, the rotational speed of the blower fan 12 is adjusted, and at least one of the opening angle and direction of the upper and lower louvers 14 and the left and right louvers 17 is adjusted.

  The wind direction time indicates a time during which at least one of the upper and lower louvers 14 and the left and right louvers 17 pauses in a predetermined direction during a swing.

  The controller 330 may make adjustments so as to be temporarily stopped for a certain wind direction time in a plurality of directions, or may determine the wind direction time based on at least one position of the upper and lower louvers 14 and the left and right louvers 17. The control unit 330 may determine the wind direction time based on the position of the person's head or the position of the foot.

  In the present embodiment, the blown air temperature is at least a set temperature, a temperature of the heat exchanger 13, a rotation speed of the compressor 19, a rotation speed of the blower fan 12, at least one opening condition and direction of the upper and lower louvers 14 and the left and right louvers 17. The temperature of the air which is adjusted by one and sent from the indoor unit 20 to the room is shown.

  The control unit 330 adjusts the temperature of the blown air according to the change in the blown air temperature due to any one of the thermo-on and the thermo-off, or the change in the suction temperature due to the change in the heat load. In addition, the control unit 330 may be configured according to the set temperature, the temperature of the heat exchanger 13, the rotational speed of the compressor 19, the rotational speed of the blower fan 12, or at least one of the upper and lower louvers 14 and the left and right louvers 17 opened or oriented. Actively adjust the air temperature.

  In this embodiment, the state of the foot refers to the presence or absence of foot detection, the position of the foot, the distance from the indoor unit 20 to the foot, the temperature around the foot, etc. It is an element that can affect the human body including the foot by changing the environment around the foot. Specifically, the state of the foot includes the surface temperature of the foot, the temperature inside the foot, and the feeling of airflow.

  According to the present embodiment, the position of the human foot can be detected more accurately by detecting the human foot from the image based on the detection result of the human head. And thereby, the comfort of a person can be improved.

  In addition, this invention is not limited to an above-described Example, Various modifications are included. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described.

  In addition, each of the above-described configurations, functions, processing units, processing procedures, and the like may be realized in hardware by designing some or all of them, for example, with an integrated circuit. Each of the above-described configurations, functions, and the like may be realized by software by interpreting and executing a program that realizes each function by the processor. Information such as a program, a table, or a file that realizes each function can be stored in a recording device such as a memory, a hard disk, or an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, or a DVD.

  Further, the control lines or information lines indicate what is considered necessary for the explanation, and not all the control lines or information lines on the product are necessarily shown. In practice, almost all the components are connected to each other.

DESCRIPTION OF SYMBOLS 10 Air conditioner 11 Remote control 12 Blower fan 13 Heat exchanger 14 Upper and lower louver 15 Front panel 16 Case base 17 Left and right louver 18 Fan 19 Compressor 20 Indoor unit

Claims (11)

An air conditioner that sends cold or warm air,
An image acquisition unit that captures an image around the air conditioner;
Detecting a human head from the captured image, and detecting a human foot within a predetermined range from the human head;
An air conditioner comprising: a control unit that adjusts air to be sent based on a position of a person's foot detected by the detection unit. The air conditioner according to claim 1,
The image acquisition unit acquires the captured first image and a second image captured in the past than the first image,
The detector is
Maintaining the relationship between the time between the time when the second image was taken and the time when the first image was taken and the movement distance that a person may move at that time;
Detecting a human head from the first image;
In the first image, a range that is more than the moving distance from the position where the human head is detected is specified in the second image, and whether or not the human head is detected from the specified range. Determine whether
If the human head is not detected as a result of the determination, the air conditioner detects the human foot using the first image and the second image. The air conditioner according to claim 1 or 2,
The image acquisition unit acquires the captured first image and a second image captured in the past than the first image,
The detector is
Extracting a region of difference in which the object represented between the first image and the second image is different;
From the extracted difference area, select a difference area that includes the head of the person detected from the first image as an area where the person is photographed,
An air conditioner that detects the human foot from the selected area. The air conditioner according to claim 3,
The detector is
In the image, having an estimation function for estimating a position of a lower end in a region where a person is photographed as a position of a person's foot within a predetermined range from the person's head,
An air conditioner that detects the position of a person's foot in the selected area using the estimation function. The air conditioner according to claim 1 or 2,
The image represents a result of measuring a physical quantity of a surface of an object around the air conditioner,
The detector is
An area including the detected human head is extracted based on a change in a physical quantity represented by the image,
Calculating an index value indicating the characteristics of the extracted region;
Based on the calculated index value, it is determined whether the extracted area represents a person,
As a result of the determination, if the extracted area represents a person, a human foot is detected from the extracted area. The air conditioner according to claim 5,
The detector is
Calculating the circularity indicating whether or not the shape of the extracted region is close to a circle as the index value;
An air conditioner that determines whether or not the extracted region represents a person based on the calculated circularity. The air conditioner according to claim 5,
The detector is
Calculating the area of the extracted region as the index value;
An air conditioner that determines whether or not the extracted region represents a person based on the calculated area. The air conditioner according to claim 5,
The detector is
Calculating the complexity indicating the complexity of the change in the physical quantity represented in the extracted region as the index value;
An air conditioner that determines whether or not the extracted region represents a person based on the calculated complexity. The air conditioner according to claim 1 or 2,
The image represents a result of measuring a physical quantity of a surface of an object around the air conditioner,
The detector is
Holding a template in the shape of the person's foot,
Based on the physical quantity represented in the image, the area including the detected person's head is extracted as an area representing the person,
In the region where the extracted person is represented, the part whose shape corresponds to the template is specified,
Extracting the identified part as a candidate area of an area in which a human foot is represented;
An air conditioner that detects the human foot from the extracted candidate area. An air conditioner according to claim 9,
Dividing the image into a plurality of blocks;
Based on a physical quantity statistic value represented by the divided block area, a physical quantity is allocated to the block,
A block image is generated using a block to which the physical quantity is assigned,
Based on the physical quantity of the detected human head area, the block image includes a plurality of blocks including blocks indicating the physical quantity of the human head as a plurality of blocks indicating the area in which the person is represented. Extracted from
In the extracted plurality of blocks, a portion whose shape corresponds to the template is specified,
The air conditioner characterized in that the specified portion is extracted as the candidate region. The air conditioner according to claim 1,
The control unit is configured to change the environment around the detected person's foot into an environment for setting the detected person's foot to a target state, based on the position of the detected person's foot, An air conditioner characterized by adjusting the air to be sent.