JP2013253717A - Air conditioner and method of controlling the air conditioner - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an air conditioner for carrying out a careful control of an air condition corresponding to the activity amount of a person staying in the room, and to provide a method of controlling the air conditioner.SOLUTION: An air conditioner includes: an imaging means 110 for taking an image of the inside of a room where an indoor unit 100 is mounted; and a control means 130 for detecting a person staying in the room based on image information input from the imaging means 110 and carrying out air conditioning control corresponding to the detected result. The control means 130 is provided with: a human body detecting part 131; a face detecting part 132; a movement amount calculation means 134 for calculating the movement amount of the person staying in the room based on the change with time of the position and size of a head part which are detected by the human body detecting part 131 or the position or size of a face which are detected by the face detecting part 132; an activity amount calculation part 135 for calculating the activity amount of the person staying in the room based on the movement amount calculated by the movement amount calculation means 134; and control changing parts 136,137 for changing the air conditioning control corresponding to the activity amount calculated by the activity amount calculation part 135.

Description

  The present invention relates to an air conditioner and an air conditioner control method.

There is known an air conditioner that detects an occupant in a room where an indoor unit is installed and reflects the detection result in air conditioning control.
For example, in Patent Document 1, a human body is identified from a moving object (moving body) that appears in the difference image by taking a difference between images input over time from the imaging means, and within a predetermined time. Describes a human body activity amount calculation device that calculates an activity feature amount (activity amount) based on the moving area of the human body.

Patent Document 2 discloses air that detects the gender, age group, and body shape of a room occupant using a face image input from an indoor unit camera (imaging means) and reflects the detection result in air conditioning control. It describes the harmonic machine.
Patent Document 3 describes an air conditioner that detects an activity state of a room occupant using image information input from an imaging unit and monitors whether or not a suspicious person has entered the room. ing.
Furthermore, Patent Literature 4 describes an air conditioner that determines the presence or absence of a resident by using an image input from an image input unit (imaging means) and performs air conditioning control according to the amount of activity of the resident. Has been.

JP-A-8-178390 JP 2010-25359 A JP 2010-71595 A Japanese Patent No. 2957378

  The human body activity amount calculation device described in Patent Document 1 detects a moving body by taking a difference between two images acquired over time, and recognizes a detected moving body that satisfies a predetermined condition as a human body. Yes. Therefore, it is impossible to accurately calculate activity feature amounts (activity amounts) because the technology cannot recognize stationary occupants and the consistency of the human body acquired over time is unclear. There's a problem.

Moreover, in the air conditioner described in Patent Document 2, for example, when it is backlit or when the occupant is facing backward with respect to the indoor unit camera (imaging means), the occupant's face is recognized. There is a problem that you can not.
Further, in the techniques described in Patent Documents 3 and 4, for example, when calculating an activity amount or the like based on a difference between two images acquired over time, the activity amount can be accurately calculated as described above. There is a problem that you can not. Further, for example, when the occupant's whole body is accurately recognized with a fine pixel interval, the amount of calculation becomes enormous. There is a problem that the manufacturing cost of the air conditioner increases because it is necessary to mount an expensive microcomputer in order to accurately grasp the movement of the occupant for each cycle time while executing such calculations.

  Then, this invention makes it a subject to provide the control method of an air conditioner and air conditioner which perform fine air-conditioning control according to the amount of activity of a room occupant.

In order to solve the above-described problem, the present invention switches the processing of the human body detection unit and the processing of the face detection unit over time or in parallel and executes the human body detected by the human body detection unit. And the face detected by the face detector based on the positional relationship between the human body and the face, and the position and size of the head detected by the human body detector, or the face detection A movement amount calculation unit that calculates the movement amount of the occupant based on the temporal change in the position and size of the face detected by the unit, and the occupancy based on the movement amount calculated by the movement amount calculation unit An activity amount calculation unit that calculates an activity amount of the person, and a control change unit that changes the air conditioning control according to the activity amount calculated by the activity amount calculation unit.
Other aspects of the present invention will be described in the embodiments described later.

  According to the present invention, it is possible to provide an air conditioner that performs fine air conditioning control according to the amount of activity of the occupants and a method for controlling the air conditioner.

It is a front view of the indoor unit of the air conditioner which concerns on 1st Embodiment of this invention, an outdoor unit, and a remote control. It is a sectional side view of an indoor unit. It is a block diagram which shows the structure containing the control means of an air conditioner. It is a flowchart which shows the flow of the process which a control means performs. (A) is explanatory drawing which shows the image imaged by an imaging means, (b) is explanatory drawing which shows the human body image detected by a human body detection means using the image of (a), (c) FIG. 5 is an explanatory diagram illustrating a state in which face detection processing is performed and the detected human body and face are associated with each other. It is a flowchart which shows the flow of a human body detection process. It is a flowchart which shows the flow of a face detection process. It is a flowchart which shows the flow of the process of matching by a labeling value. (A) is a correspondence table showing the correspondence between the occupants detected in the human body and the human body labeling values, and (b) is a correspondence table showing the correspondence between the occupants detected in the face and the human body labeling values. Yes, (c) is a correspondence table showing the correspondence between occupants and labeling values. It is a flowchart which shows the flow of a movement amount calculation process. (A) is explanatory drawing which shows the positional relationship between the occupant before a movement, a lens, and an imaging surface, (b) is the positional relationship between the occupant after a movement, a lens, and an imaging surface. It is explanatory drawing shown. It is explanatory drawing which shows the example of the image imaged by the imaging means at each time. It is explanatory drawing which shows the labeling value for every predetermined time calculated based on the image shown to Fig.12 (a)-(d), a positional information, face information, human body information, and a movement amount. It is a graph which shows the relationship between the movement amount per predetermined time of an in-room person, and activity amount. (A) is a graph which shows the relationship between the activity amount at the time of cooling, and a correction value, (a) is a graph which shows the relationship between the activity amount at the time of heating, and a correction value. It is a flowchart which shows the flow of the change process of air-conditioning control. It is a flowchart which shows the other example of the change process of air-conditioning control. It is a flowchart which shows the flow of the process which the control means of the air conditioner concerning 2nd Embodiment of this invention performs.

  Embodiments of the present invention will be described in detail with reference to the drawings as appropriate. In addition, the same code | symbol is attached | subjected to the common part in each figure, and the overlapping description is abbreviate | omitted.

<< First Embodiment >>
<Configuration of air conditioner>
FIG. 1 is a front view of an indoor unit, an outdoor unit, and a remote controller of an air conditioner according to the present embodiment. As shown in FIG. 1, the air conditioner A includes an indoor unit 100, an outdoor unit 200, and a remote controller Re. The indoor unit 100 and the outdoor unit 200 are connected by a refrigerant pipe (not shown), and air-conditions the room where the indoor unit 100 is installed by a known refrigerant cycle. The indoor unit 100 and the outdoor unit 200 transmit and receive information to and from each other via a communication cable (not shown).

The remote controller Re is operated by the user and transmits an infrared signal to the remote control receiver Q of the indoor unit 100. The contents of the signal are commands such as an operation request, a change in set temperature, a timer, an operation mode change, and a stop request. The air conditioner A performs air conditioning operations such as a cooling mode, a heating mode, and a dehumidifying mode based on these signals.
In addition, an imaging unit 110 is installed at the lower part of the center in the left-right direction of the front panel 106 of the indoor unit 100. Details of the imaging unit 110 will be described later.

FIG. 2 is a side sectional view of the indoor unit. The housing base 101 houses internal structures such as the heat exchanger 102, the blower fan 103, and the filter 108.
The heat exchanger 102 includes a plurality of heat transfer tubes 102a, and heats the air taken into the indoor unit 100 by the blower fan 103 with a refrigerant flowing through the heat transfer tubes 102a to heat or cool the air. It is configured as follows. The heat transfer tube 102a communicates with the refrigerant pipe (not shown) and constitutes a part of a known refrigerant cycle (not shown).

The left and right wind direction plates 104 are rotated by a left and right wind direction plate motor (not shown) using a rotation shaft (not shown) provided at the lower portion as a fulcrum according to an instruction from an indoor unit microcomputer (not shown). .
The vertical airflow direction plate 105 is rotated by a vertical airflow direction plate motor (not shown) around a rotation shaft (not shown) provided at both ends in accordance with an instruction from the indoor unit microcomputer.
The front panel 106 is installed so as to cover the front surface of the indoor unit 100, and is configured to be rotatable by a front panel motor (not shown) with the lower end as an axis. Incidentally, you may comprise the front panel 106 as what is fixed to a lower end.

  When the blower fan 103 shown in FIG. 2 rotates, the indoor air is taken in through the air suction port 107 and the filter 108, and the air heat-exchanged by the heat exchanger 102 is guided to the blowout air passage 109a. Further, the air guided to the blowout air passage 109a is adjusted in air direction by the left and right airflow direction plates 104 and the vertical airflow direction plate 105, and is sent to the outside from the air blowing port 109b to air-condition the room.

FIG. 3 is a block diagram showing a configuration including control means provided in the air conditioner.
The imaging means 110 is, for example, a CCD (Charge Coupled Device) camera, and is installed at the lower part of the center of the front panel 106 in the left-right direction as described above (see FIG. 1). In addition, the imaging unit 110 is installed such that the optical axis of the lens g (see FIG. 11) is directed downward by a predetermined angle with respect to the horizontal line, so that the room in which the indoor unit 100 is installed can be appropriately imaged. ing. The imaging unit 110 captures the air-conditioned room over time and outputs the captured image to the A / D converter 120 as image information.

  The A / D converter 120 is an electronic circuit that converts image information input as an analog signal from the imaging unit 110 into a digital signal and outputs the digital signal to the human body detection unit 131 and the face detection unit 132 of the control unit 130. The A / D converter 120 may be built in the imaging unit 110.

The control unit 130 controls the operation of the air conditioner A in accordance with image information input from the imaging unit 110, command signals input from the remote controller Re, sensor signals input from various sensors (not shown), and the like. Take overall control.
The storage unit 140 includes, for example, a ROM (Read Only Memory), a RAM (Random Access Memory), and the like. Then, the program stored in the ROM is read out by a CPU (Central Processing Unit) of the control means 130, expanded in the RAM, and executed.
The load 150 includes, for example, an indoor fan motor (not shown) provided in the indoor unit 100, a compressor motor (not shown) provided in the outdoor unit 200, and a vertical wind direction plate motor (not shown) installed in the vertical wind direction plate 105. And a left and right wind direction plate motor (not shown) installed on the left and right wind direction plate 104. These loads 150 are driven according to a drive signal input from the drive control unit 137 of the control means 130.

<Configuration of control means>
As shown in FIG. 3, the control means 130 includes a human body detection unit 131, a face detection unit 132, an integration processing unit 133, a movement amount calculation unit 134, an activity amount calculation unit 135, and a correction value calculation unit 136. A drive control unit 137.
The human body detection unit 131 detects a human body including the head of the occupant every predetermined time based on the image information input from the imaging unit 110, and information on the human body (hereinafter referred to as human body information). Store in the storage means 140. Incidentally, “detecting a human body including a head” includes a case where the whole body including a head (head region), a shoulder (shoulder region), and a foot (foot region) is detected.
Note that the process of the human body detection unit 131 is executed at a coarser pixel interval than the process of the face detection unit 132 described later.

The face detection unit 132 detects the face of the occupant at predetermined time intervals based on the image information input from the imaging unit 110 and stores information related to the face (hereinafter referred to as face information) in the storage unit 140. To do.
In the following, a case will be described in which human body detection processing for detecting a human body with a coarse pixel interval and processing for detecting a face with a fine pixel interval are performed while being switched over time.

The integrated processing unit 133 switches the processing of the human body detection unit 131 and the processing of the face detection unit 132 over time, and detects the human body detected by the human body detection unit 131 and the face detection unit 132. The face is associated based on the positional relationship between the human body and the face and stored in the storage unit 140.
For example, when a human body is detected at a certain time and a face is detected at a time later than the time, it is necessary to determine whether or not the detected human body and the face belong to the same person. If the integrated processing unit 133 determines that the detected human body and face belong to the same person, the integrated processing unit 133 associates each image information with a labeling value that is identification information.
Details of the association using the labeling value will be described later.

The movement amount calculation unit 134 is based on the position and size of the human body detected by the human body detection unit 131 or the temporal change in the position and size of the face detected by the face detection unit 132. The movement amount of the occupant associated with is calculated. Then, the movement amount calculation unit 134 outputs the calculated movement amount to the activity amount calculation unit 135 in association with the labeling value.
The “movement amount” means a distance estimated that the occupant has moved within a predetermined time in the real world space.

The activity amount calculation unit 135 calculates the activity amount of the occupant based on the movement amount calculated by the movement amount calculation unit 134. Then, the activity amount calculation unit 135 outputs the calculated activity amount to the correction value calculation unit 136 in association with the labeling value described above.
The “activity amount” means a metabolic amount [W / m ² ] per unit surface area of the human body, and has a positive correlation with the amount of movement described above.

The correction value calculation unit 136 changes the parameter value used for air conditioning control according to the activity amount calculated by the activity amount calculation unit 135 and outputs the parameter to the drive control unit 137.
The parameters include the set temperature of the air conditioner A, the rotational speed of the compressor motor (not shown) included in the outdoor unit 200, and the maximum amount of current supplied to the compressor motor (not shown). Including at least one. Further, the parameters include the rotational speed of an indoor fan motor (not shown), the driving of an up / down air direction plate motor (not shown), and the driving of a left / right air direction plate motor (not shown).

The drive control unit 137 receives the air conditioner A according to a remote control signal input from the remote control Re via the remote control reception unit Q and a sensor signal input from sensors such as a temperature detector (not shown). Of the load 150 (indoor fan motor, compressor motor, up / down wind direction plate motor, left / right wind direction plate motor, etc.) is controlled. Furthermore, the drive control unit 137 changes the control of the load 150 according to the correction amount input from the correction value calculation unit 136 in addition to the information.
The correction value calculation unit 136 and the drive control unit 137 correspond to a “control change unit” that changes the air conditioning control in accordance with the activity amount calculated by the activity amount calculation unit 135.

<Processing procedure of control means>
Next, an outline of a series of processes executed by the control unit 130 will be described with reference to FIG. FIG. 4 is a flowchart showing the flow of processing performed by the control means.
In step S <b> 101 of FIG. 4, the control unit 130 sets the value of n to 1 (n = 1) and stores it in the storage unit 140. Incidentally, n is a value used for counting the number of times of calculating the movement amount and the activity amount.

In step S102 of FIG. 4, the control means 130 determines whether or not a predetermined time Δt1 has elapsed since the start of processing (that is, from the time of START). The predetermined time Δt1 is a preset time (for example, several milliseconds) and is stored in the storage unit 140.
When the predetermined time Δt1 has elapsed since the start of the process (S102 → Yes), the process of the control unit 130 proceeds to step S103. On the other hand, when the predetermined time Δt1 has not elapsed since the start of the process (S102 → No), the control unit 130 repeats the process of step S102.

In step S <b> 103, the control unit 130 receives input of image information from the imaging unit 110 and stores the image information in the storage unit 140.
For example, image information corresponding to an RGB image as shown in FIG. 5A is captured by the imaging unit 110 and input to the human body detection unit 131 of the control unit 130 via the A / D conversion unit.

Next, in step S104 of FIG. 4, the control unit 130 reads out image information from the storage unit 140 and executes human body detection processing. Then, the control means 130 associates information characterizing the human body image (that is, human body information) with the human body labeling value, which is an identification symbol for each human body, and stores it in the storage means 140 (see FIG. 9A). .
Incidentally, when performing the human body detection process, the control unit 130 detects the human body of the occupant at a coarser pixel interval than the face detection process described later. Thereby, the processing load when the control means 130 performs the human body detection process can be reduced.

  For example, the occupants A, B, and C whose substantially whole body is imaged by the imaging unit 110 in FIG. 5A are rough as indicated by reference numerals A1, B1, and C1 (see FIG. 5B) by the human body detection unit 131. It is detected as a binary image. On the other hand, no human body is detected for the occupant D (see FIG. 5A) in which only a part of the body (head and shoulders) is imaged by the imaging means 110 (see FIG. 5B).

Next, in step S105 in FIG. 4, the control unit 130 reads out image information from the storage unit 140 and executes face detection processing. Then, the control means 130 associates information characterizing the face image (that is, face information) and the face labeling value that is an identification symbol for each face, and stores them in the storage means 140 (see FIG. 9B). .
For example, occupants A, B, and D whose faces are imaged by the imaging unit 110 in FIG. 5A are face images indicated by reference signs A2, B2, and D2 (see FIG. 5C) by the face detection unit 132. Detected as On the other hand, no face is detected for the occupant C facing back (see FIG. 5A) (see reference C1 in FIG. 5C).

Next, in step S106 of FIG. 4, the control means 130 uses the respective labeling values for the human body information acquired by the human body detection process (S104) and the face information acquired by the face detection process (S105). Associate. That is, when the center position of the head of each human body detected by the human body detection process and the center position of the face detected by the face detection process are within a predetermined distance, a labeling value is set between the human body and the face. Use to associate.
This makes it possible to appropriately determine whether the human body detected by the human body detection process and the face detected by the face detection process correspond to the same occupant.

Next, in step S107 of FIG. 4, the control means 130 calculates the movement amount for each occupant associated with the labeling value. The control unit 130 stores the calculated movement amount in the storage unit 140 in association with the labeling value.
In the example shown in FIG. 5, the occupants A, B, and D (see FIG. 5A) in which the face is detected correspond to the face images A2, B2, and D2 (see FIG. 5C). The amount of movement is calculated using the information.
On the other hand, the occupant C (see FIG. 5A) in which the human body was detected although the face could not be detected moved using the human body information corresponding to the human body image C1 (see FIG. 5C). Calculate the amount.

In step S108, the control unit 130 calculates an activity amount for each occupant associated with the labeling value.
Then, the control unit 130 stores the calculated activity amount in the storage unit 140 in association with the labeling value. When the value of n is 2 or more, the control unit 130 sequentially accumulates the amount of activity for each occupant associated with the labeling value (that is, calculates the sum).

In step S109, the control means 130 determines whether or not the value of n is equal to N. N is a natural number set in advance and is stored in the storage unit 140.
When the value of n is equal to N (S109 → Yes), the process of the control unit 130 proceeds to step S111. On the other hand, if the value of n is not equal to N (S109 → No), the process of the control means 130 proceeds to step S110.
In step S109, the control means 130 increments the value of n. That is, the value “1” of n set in step S101 is incremented to “2”.
In this way, the control means repeatedly executes the processes of steps S102 to S108 N times.

In step S111, the control means 130 calculates a correction value for air conditioning control. That is, the control means 130 calculates the correction value for the air conditioning control based on the activity amount for each occupant associated with the labeling value (activity amount during the time Δt1 × N).
In addition, the calculation of the correction value can use the distribution of activity amount in the room (the distribution of the left region and the right region, etc.) or the position and activity amount of a specific occupant.

  In step S112, the control unit 130 executes the air conditioning control change process using the correction value calculated in step S111. That is, the drive control unit 137 of the control unit 130 executes the air conditioning control change process using the correction value calculated based on the activity amount in addition to the above-described remote control signal and sensor signal.

<Details of each process>
(1. Human body detection process)
FIG. 6 is a flowchart showing the flow of the human body detection process.
In step S201, the control unit 130 executes binarization processing. That is, the control unit 130 reads image information input from the imaging unit 110 via the A / D converter 120 from the storage unit 140, and converts the RGB image into a binarized image. In the binarization process, for example, an RGB image is divided at a preset plurality of pixel intervals, and binarization is performed using the sum of luminance values in each region (for example, a region of 5 × 5 pixels). Run. As a result, human body recognition is performed at coarse pixel intervals to reduce the amount of calculation, thereby reducing the load on the control means 130 and speeding up the processing.

  In step S202, the control unit 130 detects the head region using the binary image information acquired by the above-described binarization process. That is, the control means 130 examines the distribution of 0 or 1 in the horizontal line from the top to the bottom of the binarized image, and corresponds to the shape (substantially circular) of the head image based on the distribution. The head region is detected depending on whether or not there exists.

  In step S <b> 203, the control unit 130 detects a foot region using information of the binary image. For example, the control unit 130 checks the distribution of 0 or 1 in the horizontal line from the bottom to the top of the binarized image, and determines whether or not there is a foot below the head acquired in step S202. To detect the foot region.

  In step S204, the control means 130 calculates the human body center. For example, the control unit 130 calculates the human body center that is the center (or center of gravity) position of the human body based on the head region acquired in step S202 and the position of the foot region acquired in step S203. Incidentally, the information on the center of the human body is acquired as position information (vertical position, horizontal position) that identifies a predetermined pixel.

  In step S205, the control unit 130 extracts a shoulder region. For example, the control unit 130 determines, for example, the number of pixels from the upper end of the head region to the lower end of the foot region and whether or not the pixel value of the position of the shoulder region estimated with respect to the head region and the foot region is 1 The shoulder region extraction process is executed.

In step S206, the control unit 130 stores the information acquired in the processing of steps S201 to S205 in the storage unit 140 as human body information in association with the labeling value for identifying the resident.
By such processing, human body information related to the human body image shown in FIG. 5B can be acquired.

(2. Face detection processing)
FIG. 7 is a flowchart showing the flow of the face detection process.
In step S301, the control unit 130 executes a filtering process to remove noise from the image information. That is, the control unit 130 reads information about the image captured by the imaging unit 110 from the storage unit 140, and removes noise from the RGB image information using, for example, wavelet transform.

In step S302, the control unit 130 detects a face area. For example, the control unit 130 obtains a color histogram for the RGB image information from which the noise has been removed, and an area in which the difference from the color histogram of the face image stored in advance in the storage unit 140 is a predetermined value or less. Is detected as a face region.
In step S303, the control means 130 detects the mouth area. That is, the control unit 130 detects the mouth region by comparing with the mouth mask image information stored in the storage unit 140 in advance in the face region detected in step S202.
In step S304, the control means 130 detects the eye area. That is, the control unit 130 detects the eye region by comparing with the eye mask image information stored in the storage unit 140 in advance in the face region detected in step S202.

In step S305, the control unit 130 stores the information acquired in the processing in steps S301 to S304 in the storage unit 140 as face information in association with the labeling value for identifying the occupant (FIG. 9B). )reference).
Through such processing, face information relating to the face image shown in FIG. 5C can be acquired.

(3. Association by labeling value)
Next, the association process using the labeling value executed by the control unit 130 will be described with reference to FIG.
In step S401 of FIG. 8, the control unit 130 sets the value of k to 1 (k = 1) and stores it in the storage unit 140. Incidentally, k is a value that is sequentially incremented when performing processing related to M human bodies detected by the human body detection processing (see S104 in FIG. 4).

Control means 130 in step S402, the distance L is equal to or less than the predetermined distance _{L 0.} Incidentally, the distance L is the distance (number of pixels) between the head center of a specific human body image corresponding to the value of k and the face center detected by the face detection process. That is, the control means 130 determines whether or not there is a face area located within a predetermined distance L ₀ from the head center of the human body image.
If there is a face area satisfying L ≦ L ₀ (S402 → Yes), the process of the control unit 130 proceeds to step S403. On the other hand, when there is no face area satisfying L ≦ L ₀ (S402 → No), the process of the control unit 130 proceeds to step S404.

In step S403, the control unit 130 stores the labeling value (face labeling value) of the face image corresponding to k = 1 in the storage unit 140 as the labeling value of the occupant. Thereby, the identity of the occupants detected by the human body detection and the face detection can be secured by associating with the labeling value. In addition, regarding the occupants recognized by both the human body detection and the face detection, the face image can be preferentially used when calculating the movement amount described later.
In step S404, the control unit 130 stores the human body labeling value in the storage unit 140 as the labeling value of the occupant. Thereby, even when the face cannot be detected, the movement amount can be calculated using the human body image.

FIG. 9A is a table showing a correspondence relationship between the occupants detected in the human body and the human body labeling values. For example, when the human body image information is acquired for the occupant C shown in FIG. 5A (see C1 in FIG. 5B), the control unit 130 converts the image information into the human body labeling value “ It is stored in the storage means 140 in association with “bbbC”.
In addition, since the human body was not detected about the occupant D shown to Fig.5 (a), the human body labeling value is not attached (refer Fig.9 (a)).

FIG. 9B is a table showing the correspondence between the occupants whose faces are detected and the face labeling values. For example, when face information is acquired for the occupant D shown in FIG. 5B (see D2 in FIG. 5C), the control unit 130 converts the face information into the face labeling value “fffD” in the face detection process described above. And stored in the storage means 140.
Note that the face labeling value is not assigned to the occupant C shown in FIG. 5A because no face is detected (see FIG. 9B).

FIG. 9C is a table showing the correspondence between the occupants and the labeling values. For example, it the head center of the occupants body images A A1 (see FIG. 5 (b)), the distance L between the face center of the face image A2 (see FIG. 5 (c)) is within the predetermined distance L ₀ (S402 → Yes in FIG. 8), the control means 130 recognizes that these are images of the same person. Then, the face labeling value “fffA” is adopted as the labeling value assigned to the occupant (see FIG. 9C).
That is, the control unit 130 uses the image information A2 (see FIG. 5C) associated with the face labeling value fffA for the movement amount calculation process described later. Thereby, the movement amount and activity amount of the occupants can be accurately calculated.

On the other hand, as shown in FIGS. 5A and 5C, no face is detected for the occupant C whose back is facing the imaging unit 110. In such a case (S402 → No in FIG. 8), the control means 130 adopts “bbbC”, which is a human body labeling value, as the labeling value assigned to the occupant C (see FIG. 9C).
That is, the control unit 130 uses the image information C1 (see FIG. 5C) associated with the human body labeling value bbbC for the movement amount calculation process described later. Thereby, even when the occupant is turning his back to the imaging unit 110 or when the face cannot be detected by backlight, the amount of activity of the occupant can be calculated.

Returning to FIG. 8, the description will be continued.
In step S405, the control unit 130 determines whether k = M. Note that the value of M is the number of detected occupants, and M = 4 in the example shown in FIG.
When k = M (S405 → Yes), the control unit 130 ends the association process using the labeling value and proceeds to the calculation of the movement amount (see step S107 in FIG. 4). On the other hand, if k = M is not satisfied (S405 → No), the control unit 130 proceeds to step S406.
In step S406, the control means 130 increments the value of k, and returns to the process of step S402.

In this way, the control means 130 preferentially adopts the face image information as the target of the activity amount calculation process while associating the human body image information and the face image with the labeling value to ensure the identity of the resident. (See FIGS. 5C and 9C). In addition, human body image information is adopted as an object of activity amount calculation processing for those for which face images could not be detected (see FIGS. 5C and 9C).
It should be noted that the above-described processing is repeated every predetermined time Δt1 (see FIG. 4), and a facial image may be used at a time that may be the same occupant and human body information may be used at other times. In this way, by switching the type of image to be processed while ensuring the identity of the occupants, it is possible to track each occupant regardless of the direction of the occupants and the lighting conditions. it can.

(3. Movement amount calculation process)
Next, the movement amount calculation process executed by the control unit 130 will be described with reference to FIG.
In step S501, the control unit 130 calculates the width P ₂ (see FIG. 11B) of the face image on the imaging surface p of the imaging unit 110. Note that for a room occupant who can detect a human body but cannot detect a face, the width of the head of the human body image on the imaging surface p is calculated.
Incidentally, the width P ₂ of the image of the face can be calculated by multiplying the width of 1 pixel per the maximum number of pixels in the horizontal direction of the image of the face.

In step S502, the control unit 130 calculates a movement amount ΔX in the left-right and up-down directions of the occupant using the amount of change in the center position of the face (or head) on the imaging surface p.
For example, the occupant moves from the state shown in FIG. 11A toward the right hand (as viewed from the imaging means 110) (ΔX toward the right and ΔY toward the near side), Consider the case of 11 (b). The movement amount ΔX at this time can be calculated using (Equation 1) shown below.
In (Expression 1), symbol H is the width of the occupant's face (or head), and is stored in advance in the storage unit 140 as a predetermined value. Further, reference numeral x ₁ is the center position of the image of the face before the movement, reference numeral x ₂ is the center position of the image of the face after the movement. Also, the amount of movement in the vertical direction can be calculated by the same method.

Next, in step S503 of FIG. 10, the control means 130 uses the amount of change in the width of the face (or head) on the imaging surface p to move the occupant in the back direction according to (Equation 2) below. The amount ΔY is calculated.
In (Expression 2), the symbol D ₁ is the distance from the occupant before the movement to the lens g, and the symbol D ₂ is the distance from the occupant after the movement to the lens g. Further, reference numeral P ₁ is the horizontal width of the image of the face of the imaging plane p before moving, reference numeral P ₂ is the horizontal width of the image of the face of the imaging plane p after movement. Reference symbol F is a distance from the lens g to the imaging surface p.

  Next, in step S504 in FIG. 10, the control means 130 uses the movement amount ΔX in the left-right direction and the movement amount ΔY in the back direction, and the actual movement amount ΔD of the occupant according to (Equation 3) shown below. Is calculated.

In this way, the control means 130 determines the movement amount ΔD of the occupant for each predetermined time Δt1 (see S102 in FIG. 4) based on the movement amount in the horizontal and vertical directions and the movement amount in the back direction. calculate. Therefore, the movement amount for each occupant can be accurately calculated.
Next, in step S505, the control unit 130 causes the storage unit 140 to store the calculated movement amount ΔD of the occupant in association with the labeling value of the occupant.

Here, calculation of the amount of movement for each occupant for each predetermined time Δt1 (see FIG. 4) will be described with reference to FIGS. Here, the predetermined time Δt1 = 1 sec.
For example, assume that the image shown in FIG. 12A is captured by the imaging unit 110 at time 9:15:00, and the image shown in FIG. 12B is captured at time 9:15:01. Incidentally, the occupant C who turned his back to the imaging means 110 in FIG. 12A turned his back to the imaging means 110 in FIG. In addition, the occupant A approaches the imaging unit 110.
When calculating the movement amount of the occupant A between the time 9:15:00 and the time 9:15:01, the control unit 130 determines the position of each face image in FIGS. 12 (a) and 12 (b). Then, based on the change in size, a movement amount of 7.259 (see FIG. 13) is calculated, and stored in the storage unit 140 in association with the labeling value NNNA (see FIG. 13).

Further, it is assumed that the image shown in FIG. 12C is picked up by the imaging means 110 at time 9:15:02, and the image shown in FIG. 12D is picked up at time 9:15:03. Incidentally, the resident B who faced the image pickup unit 110 in FIG. 12C turned his back to the image pickup unit 110 in FIG. Moreover, the occupant D who was not imaged in FIG.12 (c) is imaged in FIG.12 (d).
The control means 130 calculates the movement amount of the occupant B between the time 9:15:02 and the time 9:15:03 as follows. That is, the control unit 130 determines the amount of movement 1.B of the occupant B based on the position and size of the face image of FIG. 12C and the position and size of the head of the human body image of FIG. 268 (see FIG. 13) is calculated and stored in the storage means 140 in association with the labeling value NNNB (see FIG. 13).

  In this way, the control means 130 moves based on changes in the position and size (width) of the occupant's face that change over time, or changes in the position and size (width) of the head. Is stored in the storage unit 140 in association with the labeling value.

(4. Activity amount calculation processing)
FIG. 14 is a graph showing the relationship between the amount of movement of a person in the room per predetermined time and the amount of activity.
When calculating the amount of activity, the control means 130 sequentially calculates and accumulates the movement amount ΔD for each predetermined time Δt1 (see FIG. 4), and totals the movement amount in the predetermined time Δt1 × N (see FIG. 4). Is calculated.
Further, information corresponding to the graph of FIG. 14 is stored in advance in the storage unit 140 so that the amount of activity increases as the amount of movement per predetermined time Δt1 × N increases.
The control unit 130 calculates the amount of movement described above, calculates the amount of activity corresponding to the amount of movement, and stores it in the storage unit 140 in association with the labeling value described above.

(5. Correction value calculation processing)
Next, as an example, a case will be described in which a correction value for the set temperature of the air conditioner A is calculated using the total amount of activity of each occupant detected during a predetermined time Δt1 × N.
FIGS. 15A and 15B are graphs showing the relationship between the amount of activity and the correction value of the set temperature. The control unit 130 switches the correction value calculation method according to the operation mode (cooling operation / heating operation) received from the remote controller Re and stored in the storage unit 140.

[5-1. Correction value calculation process during cooling]
That is, in the case of cooling shown in FIG. 15A, the control unit 130 decreases the correction value of the set temperature as the calculated activity amount increases. Here, the “set temperature” is a value that is calculated by the control unit 130 using the set temperature information input from the remote controller Re and the amount of activity described above, and is referred to by the drive control unit 137.
When the activity amount is in the range of 0 or more and less than W1, there is a high possibility that there is no person, so the correction value of the set temperature is set to be high.

[5-2. Correction value calculation process during heating]
When heating, the control means 130 increases the absolute value of the set temperature correction value T8 (<0) when the calculated activity amount is 0 or more and less than W1. That is, when the amount of activity is extremely small, the control unit 130 determines that there is no occupant and reduces the set temperature in order to reduce power consumption.
In addition, it is preferable to consider not only the amount of activity but also whether the occupant is detected by the human body detection unit 131 or the face detection unit 132 in the determination of whether the occupant is present.
In the region where the activity amount is W1 or more, the control unit 130 increases the absolute value of the correction value (<0) as the activity amount value increases (that is, lowers the set temperature).

  In this way, power consumption can be reduced while performing comfortable air conditioning on the amount of activity in the room. Moreover, the air conditioner A excellent in comfort can be provided by lowering the set temperature as the amount of activity of the occupants increases.

(6. Air conditioning control change process)
[1. Wind direction control according to activity distribution]
FIG. 16 is a flowchart showing the flow of the air conditioning control change process. In addition, in FIG. 16, the case where it determines in any direction of a room | chamber left area | region, a center area | region, and a right area | region at the time of heating operation is demonstrated as an example.
In step S601, the control unit 130 determines whether or not the activity amount calculated by the activity amount calculation process is equal to or greater than a predetermined value W. The predetermined value W is a preset activity amount value. If the amount of activity is greater than or equal to the predetermined value W (S601 → Yes), the process of the control means 130 proceeds to step S602. On the other hand, when the activity amount is less than the predetermined value W (S601 → No), the process of the control unit 130 proceeds to step S606.

In step S602, the control unit 130 decreases the set temperature by a predetermined value ΔT. For example, when the amount of activity during the heating operation is within the range of W3 or more and less than W4 shown in FIG. 15B, the control means 130 adds the correction value T7 (<0) to the set temperature, and sets the set temperature to | Decrease by T7 | (= ΔT).
In step S603, the control unit 130 determines whether or not the difference between the activity amounts in the left and right regions is equal to or greater than ΔW. Note that the value of ΔW described above is preset and stored in the storage unit 140.
The amount of activity used in the determination process is the sum of the amount of activity of the occupants in the room on the left and the amount of activity of the occupants in the room on the right. It is the added value.

When the difference between the activity amounts in the left and right regions is greater than or equal to ΔW (S603 → Yes), the process of the control unit 130 proceeds to step S604. On the other hand, when the difference between the activity amounts in the left and right regions is less than ΔW (S603 → No), the process of the control unit 130 proceeds to step S605.
In step S <b> 604, the control unit 130 blows the warm air whose set temperature has been corrected toward the region with the smaller amount of activity (for example, the right region).
In step S605, the control means 130 blows warm air whose set temperature is corrected toward the center of the air-conditioned room.
Incidentally, the wind direction control in the processing of steps S604 and 605 is performed by driving a left and right wind direction plate motor (not shown) with a rotating shaft (not shown) provided at the lower part of the indoor unit 100 as a fulcrum. This is done by changing the orientation of the plate 104.

If the activity amount is less than the predetermined value W in step S601 (S601 → No), the control means 130 performs normal operation (S606).
In this way, the temperature distribution of the entire room can be made uniform by intensively blowing the hot air to the region where the amount of activity is small. Moreover, the power consumption required for air conditioning can be reduced by lowering the set temperature of the heating operation.

[2. Wind direction control according to occupant position and activity]
FIG. 17 is a flowchart illustrating another example of the air conditioning control change process. In addition, FIG. 17 demonstrates the case where wind direction control is performed according to a person's position and activity amount in the case of performing a cooling operation.
Steps S701, S702, and S704 are the same as steps S601, S602, and S606 in FIG.

In step S <b> 703, the control unit 130 performs wind direction control according to the position of the occupant and the amount of activity.
For example, when there is an occupant with an activity amount of 140 W / m ² in the right area of the air-conditioned room and there are occupants with an activity amount of 60 W / m ² in the left area, the control means 130 moves toward the right area. The control of blowing cold air for 7 seconds and then blowing cold air for 3 seconds toward the left region is repeated sequentially. As a result, wind direction control is performed according to the ratio (7: 3) of the amount of activity of the occupants located in the respective areas, so that a comfortable air-conditioning environment for the occupants can be provided.

<Effect>
According to the air conditioner A according to the present embodiment, the human body information acquired by the human body detection unit 131 and the face information acquired by the face detection unit 132 are associated with a labeling value, thereby existing in the air-conditioned room. A person in the room can be detected appropriately.
In addition, regarding the occupants recognized by both the human body detection unit 131 and the face detection unit 132, the movement amount of the occupants is accurately calculated by calculating the movement amount using the face image preferentially. can do.

Further, for the occupants who are detected by the human body detection unit 131 but not detected by the face detection unit 132, the movement amount is calculated using human body information. Therefore, even when the occupant is facing away from the imaging unit 110 or when the occupant's face cannot be detected due to backlight, the amount of movement can be calculated appropriately.
Further, in the face detection process, the face of the occupant is detected at a relatively small pixel interval, but the face area is a part of the entire human body area. Therefore, the amount of calculation can be reduced compared with the case where the whole body of the occupant is detected with a fine pixel interval, and the occupant's movement can be accurately captured.
Further, when performing the human body detection process, the human body of the occupant is detected with a coarser pixel interval than the face detection process. Thereby, the processing load when the control means 130 performs the human body detection process can be reduced.

Further, in the air conditioner A according to the present embodiment, the movement amount of the occupant for each predetermined time is calculated based on the movement amount in the left / right / up / down direction and the movement amount in the back direction. Therefore, the amount of movement for each occupant can be accurately calculated. Detailed air conditioning control is performed according to the sum of the amount of activity in the air-conditioned room or according to the distribution of the amount of activity in the air-conditioned room and the position of the resident. Therefore, comfortable air conditioning control can be realized by each occupant.

<< Second Embodiment >>
In the second embodiment, first, air conditioning control corresponding to a specific user is performed based on user information input from the remote controller Re according to a user's operation and user face image information captured by the imaging unit 110. Is different from the first embodiment. Therefore, the different parts will be described, and the description of the other parts will be omitted.

(1. User information registration procedure)
In the present embodiment, first, the control unit 130 and the remote controller Re communicate with each other, prompt the user to input user information, and store the user information input via the remote control receiving unit Q (the first storage unit). 1 storage means) 140.
Here, the user information is information regarding a specific user, such as a user name, age, sex, whether it is hot or cold. When information indicating that “registration processing is started” is input from the remote controller Re by the user's operation via the remote control receiving unit Q, the control unit 130 communicates with the remote controller Re and transmits the above-described user information. Prompt the user to enter sequentially.
Then, the control unit 130 stores the user information received via the remote control receiving unit Q in the storage unit (first storage unit) 140.

Next, the control unit 130 prompts the user to stand in front of the imaging unit 110 installed in the indoor unit 100, detects the user's face by the face image detection unit, and associates it with the registered user information. And stored in the storage means 140.
In addition, when acquiring the user information regarding a plurality of users, the control unit 130 attaches different identification symbols to the respective users.

(2. Air conditioning control according to recognition results)
FIG. 18 is a flowchart showing the flow of processing performed by the control means.
Steps S201 to S205 shown in FIG. 18 are the same as steps S101 to S105 shown in FIG.
In step S206, the control unit 130 executes personal authentication processing. That is, the personal authentication is executed by comparing one or a plurality of face images acquired in step S205 with the face images registered in the registration process. In addition, about personal authentication, since the well-known method using a feature vector etc. should just be adopted, description is abbreviate | omitted.
Steps S207 to S211 are the same as steps S106 to S110 shown in FIG.

In step S212, the control unit 130 calculates a correction value for air conditioning control corresponding to the recognized user. That is, when a user stored in the storage unit 140 is detected, the control unit 130 reads user information corresponding to the user from the storage unit 140. Then, a correction value for air conditioning control is calculated according to the user information and the amount of activity accumulated in step S209.
For example, when the user who detects the face during the cooling operation is “hot”, the control unit 130 lowers the set temperature by a predetermined value compared to the normal air conditioning control. And the direction of the left-right wind direction board 104 and the up-and-down wind direction board 105 is controlled, and cool air is blown toward the said user.
This makes it possible to execute fine air conditioning control corresponding to the registered user information.

  In the example shown in FIG. 18, the personal authentication is performed every predetermined time Δt1, but the present invention is not limited to this. For example, user authentication may be performed every predetermined time Δt1 × N (see FIG. 18) or every other predetermined time.

<Effect>
According to the air conditioner A according to the present embodiment, the user's face is imaged in advance using the imaging unit 110, and the face image information of the user is acquired. Therefore, when a registered user exists in the air-conditioned room, the user can be reliably detected.
Further, when a registered user is detected, user information input by the user is referred to, and a correction value for air conditioning control is calculated according to the user information.
Therefore, detailed air conditioning control corresponding to each registered user can be executed.

«Third embodiment»
In the third embodiment, in addition to the process in the first embodiment described above, a process for specifying the floor plan of the air-conditioned room is performed, and the air-conditioning control according to the process result is executed. Therefore, the description of the same parts as those in the first embodiment is omitted.
In general, a door (a door through which a person can enter and leave the air-conditioned room by opening and closing the door) is installed on the side facing the wall where the indoor unit of the air conditioner is installed. Therefore, the imaging unit 110 can image the door and specify the position of the door.

(1. Door area identification process)
The control unit 130 stores in the storage unit (second storage unit) 140 the position of the occupant who is detected by the human body detection process or the face detection process. And the control means 130 specifies the area | region where the human body and face of the occupant who were detected within the imaging range of the imaging means 110 are not detected as a “door area” where the door exists, and the position of the door area Information is stored in the storage means 140 (second storage means). That is, when the occupant leaves the air-conditioned room, the control unit 130 specifies an area where the occupant disappears from the field of view (imaging range) as the door area.
The door area is specified every preset period.

(2. Air-conditioning control according to the position of the door)
The control means 130 executes air conditioning control according to the open / closed state of the door in the door area and the presence / absence of a occupant in the air conditioned room.
For example, when the door is open in a cooling operation without using a timer, and there is no person in the air-conditioned room for a predetermined time, the control means 130 waits for the person in the room to disappear. The cooling operation is stopped after a predetermined time. In such a case, since it is highly likely that the occupant has forgotten to stop the operation of the air conditioner A when leaving the air-conditioned room, wasteful power consumption can be reduced by stopping the cooling operation. Can be suppressed.

Incidentally, the open state and the closed state of the door can be identified using, for example, an image pattern around the boundary of the door region. The identification process utilizes the fact that the image pattern around the door area is substantially uniform when the door is closed.
In the above example, the case where the operation is stopped when the door is open has been described, but the present invention is not limited to this. For example, when the door is open, the set temperature may be changed so as to reduce the consumption setting (increase in the case of cooling operation and decrease in the case of heating operation).

<Effect>
According to the air conditioner A according to the present embodiment, it is possible to appropriately specify the position of the door installed in the air-conditioned room by the imaging unit 110 and perform air-conditioning control according to the open / closed state of the door. That is, useless power consumption can be suppressed by stopping or weakening the air-conditioning operation when the door in the air-conditioned room is open and no occupants are present for a predetermined time.

≪Modification≫
As mentioned above, although air conditioner A concerning the present invention was explained by each embodiment, the embodiment of the present invention is not limited to these statements, and various changes etc. can be performed.
For example, in each of the above-described embodiments, the case where the human body detection process and the face detection process are switched over time is described (see steps S104 and S105 in FIG. 4), but the present invention is not limited to this.

For example, the human body detection process and the face detection process may be executed in parallel. In this case, the control means 130 determines the identity of the occupant at every predetermined time Δt2 based on the positional relationship between the human head detected by the human body detection process and the face detected by the face detection process. . Then, the movement amount and the activity amount are calculated based on the face information of each occupant (or the human body information when only the human body is detected), and the air conditioning control change process is performed in the same manner as in the first embodiment. Run.
As a result, the time interval for detecting a face or a human body can be reduced to speed up the process compared with the case where the human body detection process and the face detection process are switched over time, and the position of the occupant can be more accurately determined. Changes can be detected.

In the above-described embodiment, the case where the change process of the air conditioning control is executed by changing the set temperature has been described. However, the present invention is not limited to this.
That is, at least of the set temperature, the rotational speed of the compressor motor (not shown) included in the outdoor unit 200, and the maximum amount of current supplied to the compressor motor according to the activity amount calculated in the activity amount calculation process. One control information may be changed, and the air conditioning control may be changed using the changed control information as a target value.

  Specifically, during the cooling operation, the control means 130 increases the temperature change amount ΔT1 (≧ 0) and (Expression 5) shown below in (Expression 4) as the value of the activity amount calculated in the activity amount calculation process increases. ) Of the rotational speed change ΔR1 (≧ 0) and the maximum current change amount ΔA1 (≧ 0) of (Equation 6) are reduced.

Set temperature + ΔT1 = Changed set temperature (Equation 4)
Rotation speed of compressor motor−ΔR1 = Rotation speed of compressor motor after change (Expression 5)
Maximum current amount−ΔA1 = Maximum current amount after change (Expression 6)

On the other hand, during the heating operation, the control means 130 increases the temperature change amount ΔT2 (≧ 0) and (Equation 8) shown in (Equation 7) below as the value of the activity amount calculated in the activity amount calculation process increases. At least one change amount is increased among the speed change amount ΔR2 (≧ 0) and the maximum current change amount ΔA2 (≧ 0) of (Equation 9).
Set temperature-ΔT2 = Set temperature after change (Expression 7)
Rotation speed of compressor motor−ΔR2 = Rotation speed of compressor motor after change (Expression 8)
Maximum current amount−ΔA2 = maximum current amount after change (Expression 9)

Moreover, although the said 1st Embodiment demonstrated the example which does not estimate the age and sex of a resident, it is not restricted to this. For example, the control unit 130 may estimate the age and sex of the occupant for each occupant associated with the human body based on the positional relationship using the detection result of the face detection unit 132. In this case, the control means 130 performs the change process of air-conditioning control according to the activity amount calculated by the activity amount calculation process and the above-described estimation result.
In addition, description is abbreviate | omitted about the detail of an age and sex estimation process of a resident.
Moreover, it is good also as estimating either one among a person's age and sex, and performing the change process of air-conditioning control according to the said estimation result and activity amount.

Moreover, although each said embodiment demonstrated the case where a wind direction was changed according to distribution of activity amount (refer FIG. 16, FIG. 17), it is not restricted to this. That is, the rotational speed of an indoor fan motor (not shown) included in the indoor unit 100 may be changed according to the activity amount calculated in the activity amount calculation process. For example, when the amount of activity in the air-conditioned room during the cooling operation is large, the control unit 130 increases the amount of cold air by increasing the rotational speed of the indoor fan motor.
Further, at least one of the angle of the vertical wind direction plate 105, the angle of the left and right wind direction plate 104, and the rotation speed of the indoor fan motor may be changed.

Further, the face detection process in each of the above embodiments is performed for all areas of the captured image, but is not limited thereto. That is, the face detection process may be executed only in the area around the head of the human body detected by the human body detection process.
Thereby, the processing load when the control means 130 performs the face detection process can be reduced.

In the third embodiment, the case where the position of the door is specified as the floor plan of the room has been described. However, the present invention is not limited to this. That is, it may be determined whether the indoor unit 100 is installed on the right side or the left side of the room as viewed from the imaging unit 110, and wind direction control may be performed based on the determination result.
For example, it is determined by detecting the edge of the indoor image whether the indoor unit 100 is installed on the right side or the left side when viewed from the imaging unit 110. That is, the area of the right wall image and the area of the left wall image on the imaging surface may be compared to determine that the indoor unit 100 is installed on the smaller area side.
For example, when the indoor unit 100 is installed on the right side when viewed from the imaging unit 110, it is considered that there is a wall near the right side surface of the indoor unit 100. Therefore, the air conditioning can be performed more efficiently by blowing air toward the left side where the indoor unit 100 is not installed.

Moreover, although 2nd Embodiment demonstrated the case where the change process of air-conditioning control was performed based on the user information regarding the registered user, Furthermore, each user's action pattern was learned and the air-conditioning based on the learning result It is good also as performing control.
For example, for each user stored in the storage unit (first storage unit) 140 as user information, the position of the user in the air-conditioned room is associated with air-conditioning information including the set temperature and the wind direction when the user is in the room. Learning means (not shown) for learning and generating behavior pattern information may be provided.

  In this case, the learning unit stores the learned behavior pattern information in the storage unit 140 in association with the above-described user information. Then, when the control unit 130 recognizes the user in the face detection process, the control unit 130 reads out the behavior pattern information corresponding to the user from the storage unit 140, and executes the air conditioning control based on the behavior pattern information.

For example, when learning that a specific user is acting so that the cool air blown from the indoor unit 100 does not directly hit during the cooling operation, the learning means uses the learning result (behavior pattern information) as user information. The data are stored in the storage unit 140 in association with each other. Then, when the control unit 130 detects the user in the face detection process, the control unit 130 reads out the action pattern information corresponding to the user from the storage unit 140, so that the cold wind does not directly hit the user. The direction of the vertical wind direction plate 105 is controlled.
For example, a hierarchical neural network can be used for the learning process executed by the learning means.

In the above-described embodiments, when both the human body and the face of the same occupant are detected, the movement amount is preferentially calculated using the face image. However, the present invention is not limited to this. That is, in such a case, the movement amount may be calculated using the human body image preferentially.
Further, in each of the above-described embodiments, the case where one imaging unit 110 is installed at the lower center of the front panel 106 of the indoor unit 100 in the left-right direction has been described, but the present invention is not limited to this. For example, the imaging means may be installed on the right side or the left side of the front panel 106.
Also, a plurality of imaging means may be installed on the front panel 106.

  Further, the imaging unit 110 may include a driving unit (not shown). For example, the imaging unit 110 is connected to a shaft of a stepping motor as a driving unit, and reciprocates within a predetermined angle range by operating the motor. In this case, the motor stops for imaging for each area obtained by dividing the range of the predetermined angle into a plurality of areas, and after the imaging, the operation is resumed to direct the imaging means 110 to the next area. Note that if the time required for imaging is sufficiently short compared to the driving speed of the imaging means 110, the stepping motor may be operated continuously.

  This makes it possible to use the imaging means 110 with a narrow viewing angle and inexpensive. Also, by changing the image processing such as noise removal, face detection, human body detection, etc. according to the illuminance and luminance for each area, face detection and human body detection with higher accuracy, position and activity estimation based on them Can be performed. Furthermore, it is possible to avoid an increase in cost by mounting a plurality of inexpensive imaging means 110 having a narrow viewing angle, and to avoid an increase in the size of the indoor unit by securing an installation space for the plurality of imaging means 110. Become.

  Moreover, each above-mentioned embodiment and modification can be combined suitably. For example, the air conditioning control based on the user information and the floor plan may be executed by combining the second embodiment and the third embodiment.

A Air conditioner 100 Indoor unit 103 Blower fan 104 Left and right wind direction plate 105 Up and down wind direction plate 110 Imaging unit 120 A / D converter 130 Control unit 131 Human body detection unit 132 Face detection unit 133 Integration processing unit 134 Movement amount calculation unit 135 Activity amount Calculation unit 136 Correction value calculation unit (control change unit)
137 Drive control unit (control change unit)
140 Storage means (first storage means, second storage means)
150 load

Claims (13)

Imaging means for imaging the room in which the indoor unit is installed;
An air conditioner comprising: control means for detecting a occupant based on image information input from the imaging means and performing air conditioning control according to the detection result;
The control means includes
A human body detection unit that detects a human body including a head of a room occupant every predetermined time based on image information input from the imaging unit;
A face detection unit that detects the face of a room occupant every predetermined time based on image information input from the imaging means,
The process of the human body detection unit and the process of the face detection unit are switched over time or in parallel, and the human body detected by the human body detection unit and detected by the face detection unit The face is made to correspond based on the positional relationship between the human body and the face,
Movement for calculating the amount of movement of the occupant based on the position and size of the head detected by the human body detection unit or the temporal change in the position and size of the face detected by the face detection unit A quantity calculator;
An activity amount calculation unit that calculates an activity amount of a resident based on the movement amount calculated by the movement amount calculation unit;
An air conditioner, comprising: a control changing unit that changes the air-conditioning control according to the activity calculated by the activity calculating unit. The control means includes
The human body detected by the human body detection unit is associated with the face within a predetermined distance from the center or head of the human body by labeling using a labeling value that is identification information. The air conditioner according to 1. The air conditioner according to claim 1 or 2, wherein the face detection unit performs a face detection process on a region around the head of the human body detected by the human body detection unit. The air conditioner according to any one of claims 1 to 3, wherein the process of the human body detection unit is executed at a coarser pixel interval than the process of the face detection unit. The control changing unit controls at least one of a set temperature, a rotation speed of a compressor motor included in an outdoor unit, and a maximum current amount supplied to the compressor motor according to a calculation result by the activity amount calculation unit. The air conditioner according to any one of claims 1 to 4, wherein information is changed, and the air conditioning control is changed using the changed control information as a target value. The control change unit, during cooling operation,
As the value of the activity amount calculated by the activity amount calculation unit is larger, a temperature change amount ΔT1 (≧ 0) of (Equation 1) and a rotation speed change amount ΔR1 (≧ 0) of (Equation 2) shown below, The air conditioner according to claim 5, wherein at least one change amount among the maximum current change amount ΔA1 (≧ 0) of (Equation 3) is reduced.
Set temperature + ΔT1 = Changed set temperature (Equation 1)
Rotation speed of compressor motor−ΔR1 = Rotation speed of compressor motor after change (Expression 2)
Maximum current amount−ΔA1 = maximum current amount after change (Expression 3) The control change unit, during heating operation,
As the value of the activity amount calculated by the activity amount calculation unit is larger, a temperature change amount ΔT2 (≧ 0) of (Equation 4) shown below, a rotation speed change amount ΔR2 (≧ 0) of (Equation 5), The air conditioner according to claim 5 or 6, wherein at least one change amount among the maximum current change amount ΔA2 (≧ 0) of (Equation 6) is increased.
Set temperature-ΔT2 = Set temperature after change (Formula 4)
Rotation speed of compressor motor−ΔR2 = Rotation speed of compressor motor after change (Expression 5)
Maximum current amount−ΔA2 = maximum current amount after change (Expression 6) The control means estimates the age and / or gender of the occupant for each occupant associated based on the positional relationship between the human body and the face using the detection result of the face detection unit, The air conditioner according to claim 6 or 7, wherein the change amount is determined according to an estimation result and the activity amount. Comprising first storage means for storing user information relating to a user and a face image for each user in association with each other;
The control means includes
According to the operation of the remote control by the user, the user's face image captured by the imaging unit is stored in the first storage unit in association with the user information,
When the user stored in the storage means is detected by the face detection unit, the user information corresponding to the user is read from the first storage means, and the change is performed according to the user information and the activity amount. The air conditioner according to claim 6 or 7, wherein an amount is determined. For each user stored in the first storage means as the user information, the user's behavior in the room where the indoor unit is installed corresponds to air conditioning information including the set temperature and the wind direction when the user is in the room. With learning means to learn and generate behavior pattern information,
The learning unit stores the learned behavior pattern information in the first storage unit in association with the user information,
When the face is recognized by the face detection unit, the control means reads action pattern information corresponding to the user from the first storage means, and executes air conditioning control based on the action pattern information. The air conditioner according to claim 9. An up and down wind direction plate for adjusting the up and down direction of wind;
Left and right wind direction plates for adjusting the wind direction in the left and right direction;
An indoor fan motor for adjusting the wind speed,
The control change unit changes at least one of an angle of the up / down wind direction plate, an angle of the left / right wind direction plate, and a rotation speed of the indoor fan motor according to the activity amount calculated by the activity amount calculation unit. The air conditioner according to any one of claims 1 to 10, wherein: A second storage unit that stores a change in the position of a person in the room detected by the human body detection unit or the face detection unit with time, and floor plan information in a room where the imaging unit is installed;
The control means includes
The change in the position is read from the second storage unit, and the area where the human body and face of the occupant who has been detected within the imaging range of the imaging unit is not detected is identified as the door area where the door exists, Storing the position information of the door area in the second storage means;
The air conditioner according to any one of claims 1 to 11, wherein air conditioning control is executed in accordance with an open / closed state of the door in the door region and presence / absence of an occupant in the room. A control method for an air conditioner comprising an imaging means for imaging a room where an indoor unit is installed,
A human body detection process for detecting a human body including a head of a resident at every predetermined time based on image information input from the imaging means;
A face detection process for detecting the face of a resident at every predetermined time based on image information input from the imaging means, and switching over time or executing in parallel,
The human body detected by the human body detection process and the face detected by the face detection process are made to correspond based on the positional relationship between the human body and the face,
Movement that calculates the amount of movement of the occupant based on the position and size of the head detected by the human body detection process or the temporal change in the position and size of the face detected by the face detection process Amount calculation processing,
An activity amount calculation process for calculating an activity amount of a resident based on the movement amount calculated by the movement amount calculation process;
A control change process for changing the air conditioning control according to the activity amount calculated by the activity amount calculation process.

Images