JP5511477B2 - Air conditioner - Google Patents

Description

  The present invention relates to an air conditioner that recognizes an indoor environment based on captured image information and performs optimal air conditioning control according to the indoor environment.

  As a conventional air conditioner that recognizes the indoor environment based on the captured image information and performs optimum air conditioning control according to the indoor environment, for example, an “infrared image sensor that captures the inside of a room where air conditioning should be performed” 1, a situation determination circuit 4 that determines the situation such as the number of people, positions, movements, activity amount, clothes amount, etc. based on the image taken by the image sensor 1, and an air conditioner body 6 based on the determination result The air conditioner control circuit 5 that controls the situation determination circuit 4 is a part of the human body exposed to skin, clothing, and futon based on the temperature image from the infrared image sensor 1 during night operation. Is calculated, and an equivalent clothing amount is derived based on the calculation result ”(see Patent Document 1).

  As another example of a conventional air conditioner, “in a control device for an air conditioner controlled by a wireless remote controller, a wireless remote controller includes a transmission LED, key input conversion means, and a time signal for position detection signal output. A wireless signal receiver, a position detector for detecting a transmission direction having a plurality of light receiving elements, a signal decoding means, and a timing for detecting a position detection signal. "A reception control unit having a means and a position determination means is provided" (see Patent Document 2).

Japanese Patent Laid-Open No. 10-259942 (summary, FIG. 1) JP 63-38849 A (2 pages)

  However, the air conditioner described in Patent Document 1 needs to use an infrared light camera in addition to the CCD camera. For this reason, there are problems that the space for incorporating the camera in the air conditioner is increased and the cost is increased.

  In addition, the air conditioner described in Patent Document 2 requires a plurality of dedicated light receiving elements in the position detection unit for receiving the position detection signal transmitted from the remote controller, which increases the cost. There was a point. Moreover, even if the air conditioner described in Patent Document 2 can detect the position of the remote controller, it cannot detect who the operator is, and performs air conditioning control so that the optimum air conditioning setting is achieved according to the individual. There was a problem that it was not possible.

  The present invention has been made to solve the above-described problems, and a first object is to perform optimum air conditioning control according to the indoor environment even in a dark environment without using a plurality of cameras. It is to provide an air conditioner that can be used.

  The second purpose is to identify not only the position information of the operator but also an individual by a signal transmitted by the remote controller in a dark environment, record the individual and the air conditioning setting in association with each other, and based on the recorded air conditioning setting. An air conditioner that performs optimal air conditioning control even when no operation is performed.

An air conditioner according to the present invention includes an imaging unit capable of imaging in a band including a visible light band and a partial band of infrared light, and an image recognition unit that recognizes an indoor environment based on image information captured by the imaging unit. And an air conditioning control unit that changes the air conditioning setting based on the indoor environment recognized by the image recognition unit, and the indoor environment recognized by the image recognition unit is at least a feature of a human face or a feature of human movement , the imaging unit is shall be imaged via an infrared cut filter which transmits visible light band and near infrared band.

Air conditioner according to the present invention recognizes the image information of the imaging unit imaged via an infrared cut filter which transmits visible light band and the near-infrared light - band, performs air conditioning control based on the recognition result . Therefore, there is an effect that it is possible to perform optimum air conditioning control (for example, energy-saving air conditioning control or air conditioning control that does not impair health) even in a dark environment based on the indoor environment with an inexpensive configuration.

It is the schematic which shows the air conditioner in Embodiment 1 of this invention. It is sectional drawing which shows the imaging part in Embodiment 1 of this invention. It is a characteristic figure of the transmittance | permeability with respect to the wavelength of the infrared cut filter which comprises the imaging part in Embodiment 1 of this invention, and the sensitivity with respect to the wavelength of a sensor. It is a block diagram which shows the structure of the air conditioner in Embodiment 1 of this invention. It is a flowchart which shows the control operation of the image recognition part in Embodiment 1 of this invention. It is a discrimination function for face detection processing which constitutes an image recognition part in Embodiment 1 of the present invention. It is an image figure of the binary rectangular filter which patterns the luminance information of the image in Embodiment 1 of this invention. It is a flowchart which shows the control operation of the face recognition part which comprises the image recognition part in Embodiment 1 of this invention. It is the schematic which shows the air conditioner in Embodiment 2 of this invention. It is the schematic of the air conditioner provided with multiple light emission parts in Embodiment 2 of this invention. It is a figure which shows the irradiation range of the light emission part in Embodiment 2 of this invention, and the imaging range of an imaging part. It is the schematic including the operation part of the air conditioner in Embodiment 3 of this invention. It is the schematic of the air conditioner and operation part containing the light emission part in Embodiment 3 of this invention. It is a flowchart which shows the control operation of the image recognition part in Embodiment 3 of this invention, and an air-conditioning control part. It is a flowchart which shows the control operation of the image recognition part in Embodiment 4 of this invention. It is a flowchart which shows the control operation of the similarity determination part in Embodiment 4 of this invention. It is a figure explaining the personal recognition by the size of the person in Embodiment 4 of this invention. It is a figure explaining the posture recognition of the person in Embodiment 4 of this invention. It is a figure explaining direction recognition of a person in Embodiment 4 of the present invention.

  Hereinafter, the air conditioner of this invention is demonstrated in detail using drawing.

Embodiment 1 FIG.
FIG. 1 is a schematic diagram showing an air conditioner according to the first embodiment.
In FIG. 1, an air conditioner 1 is equipped with an imaging unit 2 (corresponding to a camera in Patent Document 1) that images a room to be air-conditioned. The mounting position of the imaging unit 2 is not limited as long as it can image a room to be air-conditioned. Therefore, the mounting position of the imaging unit 2 may be the front right side, the left side, or the center of the air conditioner 1, and may be movable so that the imaging direction can be adjusted. Further, the imaging range of the imaging unit 2 is often wide-angle so that a wide indoor area can be imaged. Therefore, even if it is mounted not on the front surface but on the lower surface of the air conditioner 1, it may be arranged in a direction in which the entire room can be imaged.

FIG. 2 is a schematic diagram illustrating a configuration of the imaging unit 2 according to Embodiment 1 of the present invention.
The imaging unit 2 includes a lens 2a, a sensor 2b, and an infrared cut filter 2c. The lens 2a collects light and delivers it to the sensor 2b, and the material thereof may be plastic or glass. Further, if the lens 2a is made of plastic, the material price is low, but productivity may be deteriorated in order to ensure performance quality. For the lens 2a, an appropriate material is selected according to the application and environment, for example, a plastic material is adjusted to ensure reliability against scratches and temperature and humidity.
The sensor 2b converts the light collected by the lens 2a into an electric signal. The sensor 2b includes a CCD and a CMOS, and the CCD has been widely used because it has high power consumption and high price but good sensitivity. On the other hand, CMOS has low power consumption and low price compared with CCD, but its sensitivity is low. However, CMOS has been able to obtain the same sensitivity as CCDs due to recent technological evolution, and has the feature of extracting information of each pixel input to the sensor, and has been introduced into many products.

The infrared cut filter 2c is disposed between the lens 2a and the sensor 2b, and has a function of cutting a partial band of infrared light from the light components collected by the lens 2a. Since the sharpness of the image is weakened by the input of infrared light, the infrared cut filter 2c is used so that only visible light can be input in order to capture a clear image. The wavelength that can be input by the sensor 2b is wider than the visible light band, and generally has sensitivity up to the infrared light band although it depends on the material characteristics used for the sensor 2b. The sensor 2b of the first embodiment also has sensitivity up to the infrared light band.
Here, the infrared light is light including near infrared light, mid infrared light, and far infrared light. The near infrared light is light having a wavelength in the vicinity of 750 nm to 2500 nm. It is assumed that the wavelength is light in the vicinity of 2500 nm to 4000 nm, and the far infrared light is light having a wavelength in the vicinity of 4000 nm to 1 mm.

FIG. 3 is a graph showing the transmittance of the infrared cut filter 2c and the sensitivity characteristic of the sensor 2b according to Embodiment 1 of the present invention.
The sensitivity characteristic A of the sensor 2b has a high sensitivity in the visible light band, and in the infrared light band having a wavelength larger than that, the sensitivity decreases, but it can be received if it is incident without entering zero. Therefore, in the imaging unit 2 according to the first embodiment, the infrared cut filter 2c having the characteristic B is used so as not to transmit a wavelength band larger than visible light. In the first embodiment, an infrared cut filter 2c that can transmit a part of infrared light is used. For example, an infrared cut filter 2c that has a characteristic of transmitting a near infrared band such as C is used. Yes. Therefore, the imaging unit 2 can capture both visible light and near infrared light. In the first embodiment, the band of light to be imaged does not have to be limited to the visible light band, and the infrared cut filter 2c may be removed from the imaging unit 2. Further, the band through which the infrared cut filter 2c can pass is not particularly limited.

FIG. 4 shows the configuration of the air conditioner 1 (more specifically, the control device for the air conditioner 1) according to Embodiment 1 of the present invention.
The air conditioner 1 according to Embodiment 1 of the present invention includes an imaging unit 2, an image recognition unit 3A, and an air conditioning control unit 4. Moreover, the air conditioner 1 is provided with the face image registration DB12 mentioned later as a recording part in this invention, for example. The imaging unit 2 outputs the captured image information to the image recognition unit 3A. Then, the image recognition unit 3 </ b> A outputs the indoor environment recognition result to the air conditioning control unit 4. And the air-conditioning control part 4 controls air-conditioning settings, such as a wind direction, an air volume, and preset temperature, based on the recognition result of 3 A of image recognition parts.

FIG. 5 is a flowchart showing the control operation of the image recognition unit 3A according to Embodiment 1 of the present invention.
The recognition by the image recognition unit 3A will be described using face recognition as an example. The indoor environment to be recognized is not limited to face information, and may be motion recognition, for example. In that case, the image recognition unit 3 </ b> A recognizes the movement that occurs in the indoor space captured by the imaging unit 2. And the air-conditioning control part 4 performs an air-conditioning setting according to situations, such as the position of the movement, a magnitude | size, or a number. This motion recognition is briefly introduced. For example, there is a method of calculating a difference in luminance value of each pixel between a background image captured in advance and a current image, and recognizing an area where the difference is equal to or greater than a preset threshold value as a moving area. Further, there is a method of calculating a difference in luminance value of each pixel between an image one frame before and an image of the current frame, and recognizing an area where the difference is equal to or greater than a threshold value as a moving area. In addition, non-patent literature "Shadow Detection by Integrating Multiple Features" , Kuo-Hua Lo and Mau-Tsuen Yang, Proceedings of 18 th International Conference on Pattern Recognition 2006, Vol.1, of the background as shown in pp 743-746 There is a method of recognizing a region in which a shadow is removed from a change region as a region where a human body moves.

Here, the description returns to the image recognition unit 3A taking face recognition as an example.
In step S1, the imaging unit 2 acquires imaging information obtained by imaging the indoor space where the air conditioner 1 is installed. In order to express this imaging information, RGB (Red red, Green green, Blue blue initial) and YUV (Y: luminance signal, U: blue component difference signal, V: red component difference signal) Expression). A characteristic of YUV is that it is easier for humans to notice changes in luminance Y than changes in hue UV, so that a large amount of information can be assigned to luminance information to efficiently represent color even with a small amount of data. Therefore, in the first embodiment, a method using the luminance Y of YUV information will be described as an example. In general, before recognizing who the face is, it is necessary to detect where the face is in the captured image.

  In step S2, the luminance Y is extracted from the photographing information YUV, and it is determined whether or not the face is detected by the face detection process in step S3, and the face is detected from the captured image. Then, the face recognition unit 5 recognizes who the face is. In the face detection process in step S3, whether or not a face is included in the input image by scanning an arbitrary partial region of the input image using an identification function as shown in FIG. 6 obtained by learning a large amount of face images. Determine whether or not. It is determined whether or not a face is included in the input image by scanning an arbitrary partial region of the input image using an identification function as shown in FIG. In the example of FIG. 6, the face image is mapped to the face image space of the discrimination function, and the soccer ball is mapped to the non-face image space, and it is determined that the soccer ball is not a face. In order to apply the input image to the discriminant function, a binary rectangular filter that patterns the luminance information of the image is used. Reference numerals 30a to 30d in FIG. 6 are binary rectangular filters, and FIG. 7 is an image diagram of the binary rectangular filter. Since this face detection process is applied to a face whose size is unknown over the entire area of the captured image, for example, if a filter is expanded in proportion to the number of assumed sizes, the data size increases, and the load of computation processing for program execution increases. Has characteristics.

FIG. 8 is a flowchart showing a control operation of face recognition unit 5 in the first embodiment of the present invention.
In step S4, the result of the face detection process is acquired, and in step S5, facial feature points are extracted and digitized. Next, in step S6, the numerical value is normalized, and in step S7, the normalized face information 11 is compared with the registered face information 13 in the face image registration DB 12 (DB: database) to collate which face. . From the collation result, the similarity is calculated in step S8, and in step S9, it is determined which face, for example, the resident himself, based on the calculated value. Then, for example, feature information such as name, age, gender, constitution, and preference is registered in advance in association with face information in the face image registration DB 12 to obtain individual feature information detected by the face detection process. Can do. Note that the face recognition in the face recognition unit 5 is a process of comparing feature points to determine who is the face detected by the face detection process, so the processing load is smaller than the face detection process.

  On the other hand, in face recognition for security such as entrance / exit management that has been commercialized in recent years, a passerby cannot pass unless the face is recognized. Facing. For this reason, the recognition unit can acquire image information including large pixels that is advantageous for the recognition process. Even when a non-imaging person closes his eyes and is not recognized, the user can take several pictures until the person is recognized.

  On the other hand, since the imaging unit 2 according to the first embodiment is installed in the air conditioner 1, the imaging unit 2 is usually a high position where the air conditioner 1 is located, and images are taken from overhead. Therefore, the imaging unit 2 has difficulty in recognizing the face, and it is difficult to identify where the person to be imaged exists in the room. Further, since the imaging unit 2 captures a wide area, the person to be imaged may be small and face recognition is difficult. In addition, the conventional air conditioner has a problem that it cannot obtain luminance information necessary for recognition, particularly when the room is dark. Therefore, in the air conditioner 1 according to the first embodiment, the infrared cut filter 2c is removed as described above, or the infrared cut filter 2c having a characteristic of transmitting a part of the infrared light band as shown in FIG. used. Thereby, by increasing the information incident on the sensor 2b, that is, by increasing the sensitivity, necessary luminance information can be acquired even in the room where the air conditioner 1 is installed. By transmitting infrared light, the sensor 2b also inputs information other than the visible light band. Note that when the infrared light band is incident, color accuracy is deteriorated, but information such as shape and position is acquired correctly.

  In the above description, personal face recognition has been described as an example. However, an object to be recognized as an indoor environment may be not only a face but also a body part such as a limb or an object such as a sofa or a TV. For example, air-conditioning control can be performed by gesture operation by recognizing hand movement, and if the position of a wall or TV is known, air-conditioning in a useless direction that does not require air-conditioning can be eliminated.

  As described above, the air conditioner 1 according to the first embodiment enables the imaging unit 2 to perform imaging in the near infrared band. Therefore, it is possible to correctly acquire information such as the shape and position of the imaging target and recognize the indoor environment even in a dark environment with an inexpensive configuration. And the air-conditioning control part 4 can perform the optimal air-conditioning control according to the indoor environment using the recognition result. For example, if the air-conditioning control unit 4 obtains information about the preference of the individual air-conditioning setting and the position of the individual, it can be determined whether the reaching wind speed at that position or the ambient temperature at that position is the preferred air-conditioning setting. For example, when a person who hates wind and has a favorite temperature of 25 degrees is in the room, the person hits the wind and the ambient temperature is 27 degrees. In that case, the air-conditioning control unit 4 can create a comfortable environment by blowing air so that the person does not hit the wind and adjusting the temperature to be 25 degrees by lowering the set temperature.

  Moreover, since the air conditioner 1 in this Embodiment 1 can perform required air conditioning in a required space, useless air conditioning can also be eliminated and it can also operate | move with energy saving. In addition, the air conditioning control unit 4 obtains information on the position and posture of the person. For example, when the person is lying down, the air conditioning control unit 4 blows so as not to directly apply cold air to the body. Divide and reduce air volume. In such a case, since the amount of human activity is small, the air-conditioning control unit 4 can also prevent health deterioration such as prevention of falling asleep by raising the set temperature. Furthermore, by using motion recognition as described above, air conditioning can be controlled by gesture operation, so that it is possible not to mount an operation unit (corresponding to the remote control in the prior art document 2) on the air conditioner 1.

Embodiment 2. FIG.
In the second embodiment, the air conditioner 1 according to the first embodiment to which the light emitting unit 6 is added will be described. In the second embodiment, items that are not particularly described are the same as those in the first embodiment, and the same functions and configurations are described using the same reference numerals.

  FIG. 9 is a schematic diagram illustrating an example of the air conditioner 1 according to Embodiment 2 of the present invention, and is a schematic diagram illustrating the air conditioner 1 in which a light emitting unit 6 is added to the air conditioner 1 of FIG. . FIG. 10 is a schematic diagram of the air conditioner 1 including a plurality of light emitting units 6, and FIG. 11 is a diagram illustrating an irradiation range of the light emitting unit 6 and an imaging range of the imaging unit.

  The configuration of the control of the air conditioner 1 in the second embodiment is the same as that in FIG. 4 described in the first embodiment, and the configuration of the imaging unit 2 is the same as that in FIG. 2 described in the first embodiment. Moreover, the air conditioner 1 of this Embodiment 2 adds the light emission part 6 to the air conditioner 1 of FIG. The light emitting unit 6 attached to the air conditioner 1 emits light in an infrared light band that can be transmitted by the infrared cut filter 2c, and the light emission timing and light emission intensity may be adjustable. If the characteristic of the infrared cut filter 2c is a characteristic that transmits infrared light as shown in FIG. 3, the light emitting unit 6 may be the same light emitting element as that used in the operation unit of the air conditioner 1, If the near-infrared light emitting element is mounted for communication in the air conditioner 1 capable of bidirectional communication, it may be used.

  Further, the number of light emitting units 6 is not necessarily one, and may be plural as shown in FIG. For example, when there are a plurality, some of them may face the ceiling. The arrangement of the light emitting unit 6 is desirably the front or lower part of the air conditioner 1, and the light emitting direction is desirably matched to the imaging direction of the imaging unit 2.

  In addition, as shown in FIG. 11, if the imaging range and the light emission range overlap, it is not always necessary to be side by side with the imaging unit 2. If the imaging unit 2 is provided such that the imaging range moves, the light emitting unit 6 may be movable so that the emission range is changed in accordance with the movement.

Next, the operation of the air conditioner 1 will be described.
The light emitting unit 6 usually emits infrared light only at the timing of imaging when it is determined that the imaging unit 2 is dark. For example, when the illuminance in the room is 10 Lx or less, the light emitting unit 6 emits infrared light in accordance with the imaging timing of the sensor 2b of the imaging unit 2. If the sensitivity of the sensor 2b is high, the threshold value at which the light emitting unit 6 emits light need not be 10Lx. The threshold value may be 0.1 Lx, for example, according to the sensor sensitivity. Here, if the image captured by the image capturing unit 2 when the infrared light is irradiated appears to be blurred white, it can be determined that the output of the infrared light is strong. Adjustments may be made. In addition, when the light emitting unit 6 is equipped with a plurality of light emitting elements, the number of light emitting elements may be reduced. The light emitting unit 6 may adjust the light emitting direction, such as indirectly illuminating the imaging range in a direction different from the imaging direction of the imaging unit 2, for example, the ceiling direction.

In general, since a person rarely moves in a room in a dark environment, it is not necessary to recognize an image of the entire room in which the air conditioner 1 is installed. In addition, when the lens 2a or the sensor 2b of the imaging unit 2 has an inexpensive configuration, the captured image may be distorted or the image may not be recognized over the entire imaging range because the number of pixels is small. Therefore, the light emitting unit 6 may irradiate only the range in which the face can be recognized, for example, in the imaging range of the imaging unit 2.
In this way, the image recognition process described in the first embodiment is performed based on the image information captured even in a dark environment by irradiating the imaging target with the light of the light emitting unit 6.

  As described above, the second embodiment has the same effect as that of the first embodiment, but also has the following effect. Since the image recognition unit 3A is equipped with the light emitting unit 6 and can acquire image information even in a dark environment, the indoor environment in a dark environment can be more easily recognized than in the first embodiment. For example, even when the sensor 2b having low sensitivity is used, the indoor environment can be recognized in a dark environment. Therefore, the air conditioner 1 can determine the state of the air-conditioned room regardless of the environment with an inexpensive configuration, and can perform optimal air-conditioning control. In addition, the imaging unit 2 can recognize an indoor environment without giving an unpleasant feeling like a camera flash because it can capture an image without being noticed by a person using infrared light. Since the light emission of the light emitting unit 6 is normally dark and only at the timing when the image pickup unit 2 takes an image, there is no unnecessary light emission and the operation can be performed with energy saving. In addition, since imaging in a dark environment is made possible further than in the first embodiment, the air conditioner 1 installed at a position with a good view can be used for safety confirmation and security purposes.

Embodiment 3 FIG.
In the third embodiment, an air conditioner 1 including an operation unit 7 that changes air conditioning settings or displays an operation state will be described. 12 is a schematic diagram of the air conditioner 1 according to Embodiment 3 of the present invention, FIG. 13 is a schematic diagram of the air conditioner 1 including the light emitting unit 6 and the operation unit 7, and FIG. 4 is a control flow of the image recognition unit 3B and the air conditioning control unit 4 in Embodiment 3 of the present invention. In Embodiment 3, items that are not particularly described are the same as those in Embodiment 1 or Embodiment 2, and the same functions and configurations are described using the same reference numerals.

  The operation unit 7 includes a communication unit 9 that transmits the content of the air conditioning setting as infrared light. The communication unit 9 is conventionally mounted on the operation unit 7 in order to change the air conditioning setting of the air conditioner 1. A near-infrared light emitting element may be used. Moreover, in this Embodiment 3, the light conditioner 8 is mounted in the air conditioner 1, the light receiver 8 receives the light emitted from the communication unit 9, and the air conditioning controller 4 is based on the received contents. The air conditioning setting information operated by the operation unit 7 is recognized. And the light which the communication part 9 emits is infrared light which can permeate | transmit the infrared cut filter 2c of the imaging part 2, and it is good also as adjustment of light emission timing and light emission intensity. When light emitted from the communication unit 9 or reflected light thereof enters the sensor 2b mounted on the imaging unit 2, the imaging unit 2 images infrared light emitted from the communication unit 9.

  Further, the air conditioner 1 may include a light emitting unit 6 as shown in FIG. 13, and the light emitting unit 6 may be a near infrared light emitting element that has been conventionally mounted to communicate with the communication unit 9. Further, the number of the communication units 9 is not necessarily one as in the light emitting units 6 of the air conditioner 1, and may be plural. When there are a plurality, it is desirable to arrange them so that their light emitting directions are different. FIG. 12 is a diagram in which the operation unit 7 is provided in the first embodiment. FIG. 13 is a diagram in which the operation unit 7 is provided in the second embodiment.

Next, the operation of the air conditioner 1 will be described.
When the operation unit 7 is operated, the air conditioning setting is transmitted from the communication unit 9 as a near infrared light signal. And the light-receiving part 8 mounted in the air conditioner 1 receives the near-infrared light signal. And the air-conditioning control part 4 changes an air-conditioning setting based on the content of the received signal. The imaging unit 2 captures an image of the entire room, and simultaneously captures near-infrared light transmitted from the communication unit 9 when the operation unit 7 is operated. The image recognition unit 3B identifies the position of the operation unit 7 and recognizes the operator who has operated the operation unit 7.

FIG. 14 is a control flow of the image recognition unit 3B and the air conditioning control unit 4 according to Embodiment 3 of the present invention.
An image recognition unit 3B in FIG. 14 is provided in place of the image recognition unit 3A included in the air conditioner 1 in FIG. 4 described in the first embodiment. Steps S1 and S2 are the same as the processing in FIG. 5 of the first embodiment. In step S11, a change in luminance due to incidence of near infrared light received by the light receiving unit 8 is detected. Regarding the detection of the luminance change, the movement recognition has been described in the first embodiment, but the luminance change may be detected in the same manner as the recognition method. For example, the method may be a method of recognizing by the difference between the background image and the current image, or a method of recognizing by the difference between the image one frame before and the image of the current frame. In step S12, it is detected at which position in the imaging range the communication unit 9 emits light. In order to confirm that the light detected in step S12 is the light emitted from the communication unit 9, for example, there is the following method. The recording unit has in advance information such as the size and emission range of the near-infrared light signal transmitted by the communication unit 9, that is, the luminance value and the luminance change range at the time of transmission of the communication unit 9 imaged by the imaging unit 2. Keep it. In step S11, if light having a characteristic similar to that characteristic is recognized in the area where the luminance has changed, it can be determined that the operation unit 7 and the communication unit 9 are present.

  There are cases where the imaging unit 2 misses the light emission of the communication unit 9 because the imaging timing of the imaging unit 2 and the transmission timing of the communication unit 9 do not match. In order to prevent this, the light emission time of the communication unit 9 may be lengthened according to the imaging interval of the imaging unit 2 (generally referred to as a frame rate). In order to lengthen the light emission time, the content of the air conditioning setting of the operation unit 7 is transmitted from the communication unit 9 as usual, and separately from this, the communication unit 9 emits infrared light used for specifying the position of the operation unit 7. Since the emitted light is not related to the communication of the air-conditioning setting, if the emission is continued at least during the imaging interval of the imaging unit 2, it is not missed. Further, when the light receiving unit 8 receives the content of the air conditioning setting of the operation unit 7, the imaging unit 2 may eliminate the missed shooting by increasing the imaging interval for a predetermined time in preference to other processing.

  Moreover, when the light conditioner 6 is provided in the air conditioner 1 as shown in FIG. 13, the light emitting unit 6 according to the timing at which the light receiving unit 8 receives the near infrared light signal transmitted by the communication unit 9. And the imaging unit 2 takes an image. By doing so, only when the operation unit 7 is operated, that is, when a person is surely present in the room, the imaging unit 2 can image the room even in a dark environment. Furthermore, the air conditioner 1 may include the light emitting unit 6 and the light receiving unit 8, and the operation unit 7 may include the communication unit 9 and the light receiving unit to realize bidirectional communication. In that case, the light emitting unit 6 of the air conditioner 1 may request the operation unit 7 to emit light so that the communication unit 9 repeatedly emits light until the position of the operation unit 7 can be detected.

  In the image recognition unit 3B, the face detection process in step S3 is performed on the captured information after the processes in steps S1 and S2. This face detection process is the same as step S3 in FIG. 5 of the first embodiment. The face recognition unit 5 performs face recognition processing using the detected result. In step S10 of the air conditioning control unit 4, the contents of the air conditioning setting of the operation unit 7 and the image information of the operator are recorded in association with the face image registration DB 12 used in the face recognition described in the first and second embodiments. To do. As described in the first embodiment and the second embodiment, the image recognition unit 3B uses the recorded information so that the air-conditioning control unit 4 can use the above-described recorded information even if the operator does not perform an operation at night, for example. It can be automatically changed to the air conditioning setting preferred by the operator.

  As described above, the third embodiment has the same effect as that of the first embodiment, but also has the following effect. The image recognition unit 3B can recognize the position of the operator who has set the air conditioning by accurately and accurately detecting the position of the operation unit 7 even in a dark environment. Therefore, even when there are a plurality of persons in the air-conditioned space, it is possible to identify an individual based on the position of the operation unit 7 and record the personal preference of the air-conditioning setting. And the air-conditioning control part 4 can be changed to the personal preference air-conditioning setting, without requiring operation by using the recorded information. Furthermore, air conditioning can be performed only in a space where an individual exists. As described above, the air conditioner 1 according to the third embodiment can perform optimum air conditioning control with respect to the indoor environment and can be operated with energy saving.

Embodiment 4 FIG.
In the first to third embodiments, the method for recognizing a person existing in the environment where the air conditioner 1 is installed has been described. However, there is a case where the indoor environment cannot be recognized in a dark environment because the distance from the imaging unit 2 to the person is far away or the light emission unit 6 does not emit light sufficiently. Therefore, in the fourth embodiment, a method for recognizing an individual even when the face recognition unit 5 cannot recognize the face based on the image captured by the imaging unit 2 will be described.

  FIG. 15 shows an image recognition unit 3C according to the fourth embodiment, which is used instead of the image recognition unit 3A of the air conditioner shown in FIG. FIG. 16 shows a similarity determination unit 10 which is a part of the image recognition unit 3C of FIG. The similarity determination unit 10 in FIG. 16 is similar to the configuration of the face recognition unit 5 in FIG. 8 but is characterized in that the object to be recognized is not only the face but also the body. The personal feature registration DB 17 stores not only the face image but also personal features such as a recognized face, body image, action pattern, and preference for air conditioning setting. The personal feature registration DB 17 is used instead of the face image registration DB 12 in the configuration of the air conditioner of FIG. In the fourth embodiment, items that are not particularly described are the same as those in the first to third embodiments, and the same functions and configurations are described using the same reference numerals. FIG. 17 is a diagram for explaining recognition of the size of a person in the fourth embodiment, FIG. 18 is a diagram for explaining posture recognition in the fourth embodiment, and FIG. 19 is a diagram in the fourth embodiment. It is a figure explaining direction recognition.

Next, the operation will be described.
The similarity determination unit 10 recognizes a face, a position, and the like, and also recognizes body characteristics such as the size, posture, and orientation of the person who has the face. For example, size recognition will be described as a body feature recognition method. If it is possible to recognize a movement, that is, a range where there is a change in luminance, it is understood that there is a high possibility that a person exists in that range. Therefore, it can be estimated that there is a person at the position of the range, and the size of the range is the size of the person. Then, as shown in FIG. 17, it can be determined that an adult (male) 21 or an adult (female) 20 is an adult (male) 21 if the range where the luminance change is large and a child 19 is smaller. Note that there is the method described in the first embodiment for motion recognition.

  The determination process in the similarity determination unit 10 may be erroneously recognized, for example, when it is hidden by the table or only a part of the body is moved when it is determined only by movement. Therefore, the recognition accuracy can also be improved by extracting a face first by the face detection process in step S3 and recognizing that the range where the face exists is a human range in the range with movement. Furthermore, by continuously recognizing the range of motion not only at a certain time but at an arbitrary interval, it is also possible to recognize the movement of a part of the body hidden in the object, such as a foot hidden on the sofa, over time. This may improve the recognition accuracy.

  As for the posture recognition, it can be determined that the posture is horizontal if the range of the movement is horizontally long by the recognition of the movement in the same manner as the size. Also, recognition accuracy can be improved by recognizing the face of the posture. For example, if it is recognized that the eyes and mouth are arranged horizontally as shown in FIG. 18, it can be recognized as a sleeping person 23. Further, it can be recognized as a sitting person 22 or a standing person 24 depending on the height of the face. As for the direction, if both eyes can be recognized as shown in FIG. 19, the person 26 is facing the front, the person 25 facing left when only the right eye is recognized with respect to the nose, and the person 27 facing right when only the left eye is recognized. Can be recognized.

  Moreover, you may recognize the characteristics, such as a man and woman, age, by the information of a hairstyle or clothes. In the first embodiment, the method of using the luminance Y in the YUV of the captured image has been described. However, color information U and V other than Y may be used. For example, it is possible to identify an individual with a favorite outfit from registration information of the outfit color. In particular, in general homes, there are many adult men, adult women, and children who are present in the indoor environment where the air conditioner 1 is installed. Therefore, an adult or a child can be discriminated by the size of the body, and a man and a woman can be roughly discriminated by the hairstyle, body shape, and clothes.

  In this way, the body features such as size, posture and orientation other than the face, the preference of the air conditioning setting of the person with the face, the operation history and the feature information such as the existing position are stored in the personal feature registration DB 17 of FIG. Keep it. If the feature is updated, the information in the personal feature registration DB 17 is updated. As described above, in a dark environment where the distance from the imaging unit 2 to the imaging target is far away, out of the face recognition range, or when the light emitting unit 6 is insufficiently lit, the captured image becomes coarse and the face cannot be recognized. There is a case. Moreover, since the installation position of the air conditioner 1 is overhead, the face recognition unit 5 may not be able to recognize the face because the room is imaged from above. However, as described above, there is also body feature information that can be recognized by motion recognition, such as body size and posture, even if the face cannot be recognized.

Therefore, as shown in FIG. 15, when the light / dark determination unit in step S13 determines that the environment is a dark environment, similarity determination as shown in FIG. 16 is performed.
First, imaging information is acquired in step S14. The processing of the next step S15 and step S16 is the same as the processing of step S5 and step S6 in FIG. 8, and only the object to be recognized is different. In step S17, the recognized body information 16 and the registered body information 18 in the personal feature registration DB 17 are collated by body authentication processing. In step S18, the similarity is calculated using the collation result, and in step S19, if the similarity is equal to or greater than a predetermined value, the person is identified. The information stored in the personal feature registration DB 17 may be not only a person but also an object such as a table, a chair, or a sofa. In that case, it may be determined whether or not the object is based on information that can be partially recognized even in a dark environment. Then, based on the recognized indoor environment, it is possible to determine what kind of setting the air conditioning setting can optimally control the air conditioning of the person in the room and the air conditioning space in which the person exists.

  As described above, the fourth embodiment has the same effects as those of the first embodiment, but in addition to the effects described below. By determining similarity between information recognizable from image information larger than the face, such as the size and posture of a person, and physical information recorded in advance, an individual in a dark environment or a place away from the imaging unit 2 can be identified. Therefore, the air conditioning control unit 4 can adjust the air conditioning settings using the air conditioning settings associated with the individual registered in the personal feature registration DB 17. A person in the room where the air conditioner 1 is installed can obtain an optimum air conditioning environment in the indoor environment where the air conditioner 1 is installed without performing an air conditioning setting operation.

  DESCRIPTION OF SYMBOLS 1 Air conditioner, 2 imaging part, 2a lens, 2b sensor, 2c infrared cut filter, 3A image recognition part, 4 air-conditioning control part, 5 face recognition part, 6 light emission part, 7 operation part, 8 light reception part, 9 communication part 10 similarity determination unit, 11 face information, 12 face image registration DB, 13 registered face information, 16 body information, 17 personal feature registration DB, 18 registered body information, 19 children, 20 adults (female), 21 adults (male) , 22 Sitting person, 23 Sleeping person, 24 Standing person, 25 Person facing left, 26 Person facing front, 27 Person facing right, 30a-30d Binary rectangular filter.

Claims (13)

An imaging unit that images a room in a band including a visible light band and a partial band of infrared light;
An image recognition unit for recognizing an indoor environment based on image information captured by the imaging unit;
An air conditioning control unit for changing the air conditioning setting based on the indoor environment recognized by the image recognition unit;
Equipped with a,
The indoor environment recognized by the image recognition unit is at least a feature of a human face or a feature of human movement,
The imaging unit, an air conditioner which is characterized that you captured via an infrared cut filter which transmits visible light band and near infrared band. The imaging unit
A lens that collects light,
A sensor that converts the collected light into an electrical signal ;
With
The air conditioner according to claim 1, wherein the infrared cut filter is provided between the lens and the sensor . Including at least one light emitting section that emits infrared light;
The air conditioner according to claim 1 or 2, wherein the light emitting unit emits light according to a timing at which the imaging unit captures an image. The air conditioner according to claim 3, wherein at least one of the light emitting units emits light in a ceiling direction and indirectly irradiates an imaging range .   5. The air conditioning according to claim 3, wherein at least one of a light emission output and a light emission direction of infrared light emitted from the light emitting unit is adjusted according to an environment in which the imaging unit images. Machine. An operation unit having a communication unit for transmitting the content of the air conditioning setting as infrared light;
A light receiving unit for receiving infrared light transmitted by the communication unit;
A recording unit for recording an indoor environment and an air conditioning setting associated with the indoor environment;
With
The imaging unit images the infrared light transmitted by the communication unit,
The image recognizing unit specifies a position of the operation unit and an operator based on image information captured by the imaging unit,
The air conditioner according to claim 1 or 2, wherein the air conditioning control unit records the specified operator and the content of the air conditioning setting received by the light receiving unit in the recording unit in association with each other. . An imaging unit that images a room in a band including a visible light band and a partial band of infrared light;
An image recognition unit for recognizing an indoor environment based on image information captured by the imaging unit;
An air conditioning control unit for changing the air conditioning setting based on the indoor environment recognized by the image recognition unit;
An operation unit having a communication unit for transmitting the content of the air conditioning setting as infrared light;
A light receiving unit for receiving infrared light transmitted by the communication unit;
A recording unit for recording an indoor environment and an air conditioning setting associated with the indoor environment;
With
The imaging unit images the infrared light transmitted by the communication unit,
The image recognizing unit specifies a position of the operation unit and an operator based on image information captured by the imaging unit,
The air conditioning control unit records the recorded operator in the recording unit in association with the specified operator and the content of the air conditioning setting received by the light receiving unit. The air conditioner according to claim 6 or 7 , wherein the communication unit emits infrared light used for position determination separately from transmission of air conditioning setting contents. Including at least one light emitting section that emits infrared light;
If the imaging environment is less than the predetermined light amount,
The light emitting unit, when the imaging unit is the imaging, according to any one of claims 6 to claim 8, characterized in that irradiates infrared light in the range including the position of the operator Air conditioner. A recording unit for recording the indoor environment and the air conditioning setting associated with the indoor environment;
The air conditioning controller
Record the indoor environment and the air conditioning setting associated with the indoor environment in the recording unit,
If the recognized indoor environment is similar to the recorded indoor environment,
The air conditioner according to any one of claims 1 to 5, wherein the air conditioner is changed to an air conditioning setting recorded in association with the indoor environment. The air conditioning controller
Record the indoor environment and the air conditioning setting associated with the indoor environment in the recording unit,
If the recognized indoor environment is similar to the recorded indoor environment,
The air conditioner according to any one of claims 6 to 9 , wherein the air conditioner is changed to an air conditioning setting recorded in association with the indoor environment. The image recognition unit
A light / dark determination unit that determines whether the imaging environment of the imaging unit is a bright environment or a dark environment;
A similarity determination unit that determines the degree of similarity between the indoor environment recognized by the image recognition unit based on image information of the imaging unit and the indoor environment recorded in the recording unit;
With
When the imaging unit images in a dark environment,
The similarity determination unit determines a similarity between the indoor environment recognized by the image recognition unit based on image information of the imaging unit and the indoor environment recorded in the recording unit,
If the similarity between the two is greater than or equal to a predetermined value,
The air conditioning controller
It changes to the air-conditioning setting linked | related with the said indoor environment currently recorded on the said recording part, The air conditioner of Claim 10 or Claim 11 characterized by the above-mentioned. An imaging unit that images a room in a band including a visible light band and a partial band of infrared light;
An image recognition unit for recognizing an indoor environment based on image information captured by the imaging unit;
An air conditioning control unit for changing the air conditioning setting based on the indoor environment recognized by the image recognition unit;
A recording unit for recording an indoor environment and an air conditioning setting associated with the indoor environment;
With
The air conditioning controller
Record the indoor environment and the air conditioning setting associated with the indoor environment in the recording unit,
The image recognition unit
A light / dark determination unit that determines whether the imaging environment of the imaging unit is a bright environment or a dark environment;
A similarity determination unit that determines the degree of similarity between the indoor environment recognized by the image recognition unit based on image information of the imaging unit and the indoor environment recorded in the recording unit;
With
When the imaging unit images in a dark environment,
The similarity determination unit determines a similarity between the indoor environment recognized by the image recognition unit based on image information of the imaging unit and the indoor environment recorded in the recording unit,
If the similarity between the two is greater than or equal to a predetermined value,
The air conditioning controller
It changes to the air-conditioning setting linked | related with the said indoor environment currently recorded on the said recording part, The air conditioner characterized by the above-mentioned.

Images