JP2017053603A - Air conditioner - Google Patents

Abstract

PROBLEM TO BE SOLVED: To properly adjust indoor air, by accurately grasping structure of a room.SOLUTION: An air conditioner includes: a photographing unit configured to photograph a plurality of images in a room; a storage unit configured to store the plurality of photographed images; a conversion unit configured to specify a region where strain is removed from the plurality of photographed images, and image-convert the specified region; a detection unit configured to detect features in the room on the basis of an image containing the region having been subjected to image conversion; and a control unit configured to adjust an air amount fed to the room or a wind direction on the basis of the features of the room detected by the detection unit.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an air conditioner.

  There is provided an air conditioner that adjusts the temperature or humidity of indoor air so that the room is comfortable for a person. Conventional air conditioners adjust the indoor air to a temperature or humidity specified by the user.

  In addition, for example, a space recognition device that detects an obstacle in a shooting target region from a camera image and an air conditioner using the space recognition device have been proposed (for example, see Patent Document 1).

JP 2010-190432 A

  In the room, a door, a bran, a shoji screen, etc. for people to enter and exit is installed. When the room partition is opened, ventilation is performed between the room where the air conditioner is installed and the adjacent room or hallway, and the temperature or humidity of the room changes.

  Conventional air conditioners do not maintain accurate information about room partitions. For this reason, even if the conventional air conditioner changes to a state where the partition of the room is opened, it cannot control the strength of the wind direction and the output according to the change of the partition. As a result, there was a place where the temperature etc. could not be adjusted near the open partition (when the air conditioner was in cooling operation, a heat pool) was generated, so the air in the entire room could not be adjusted to the desired temperature or humidity. .

  Moreover, the shape and size of the room are not constant. Conventional air conditioners do not hold accurate information regarding the floor plan indicating the shape and size of the room. For this reason, the conventional air conditioner cannot adjust the air according to the room layout, for example, by increasing the output to the farthest part from the air conditioner in the room. As a result, there are places in the room where the temperature or the like cannot be adjusted, and the air in the whole room cannot be adjusted to a desired temperature or humidity.

  An object of the present invention is to provide an air conditioner that can accurately detect room features that are not captured by a camera with a narrow viewing angle, and that prevents the accumulation of heat.

  A typical example of the present invention is as follows. That is, an air conditioner that identifies an imaging unit that captures a plurality of indoor images, a storage unit that stores the captured images, and an area from which distortion is removed from the captured images And a conversion unit that converts the image of the identified area, a detection unit that detects a feature of the room based on an image including the image-converted area, and a feature of the room detected by the detection unit. And an air conditioner characterized by comprising a controller for adjusting the amount or direction of air sent to the room.

  According to an embodiment of the present invention, even when an image captured by a wide-angle camera is used, it is possible to accurately detect room features that are not captured by a camera with a narrow viewing angle, thereby preventing the occurrence of thermal accumulation. .

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

It is a front view which shows the indoor unit of the air conditioner of Example 1, an outdoor unit, and a remote control. It is explanatory drawing which shows the sectional side view of the indoor unit of Example 1. FIG. 6 is an explanatory diagram illustrating an imaging range by an imaging unit according to Embodiment 1. FIG. It is explanatory drawing which shows the temperature detection range by the temperature detection part of Example 1. FIG. 6 is an explanatory diagram illustrating an imaging range by an imaging unit including the wide-angle lens of Example 1. FIG. It is explanatory drawing which shows the method of rotating the imaging part and temperature detection part of Example 1. FIG. It is explanatory drawing which shows the other example of the position where the imaging part and temperature detection part of Example 1 are arrange | positioned. It is a functional block diagram which shows the outline | summary of the process which detects the partition of Example 1 and air-conditions a room | chamber interior. 6 is a flowchart illustrating processing for detecting a partition when the imaging unit according to the first embodiment rotates. FIG. 5 is an explanatory diagram illustrating functional blocks of a processing unit of the space detection unit and image processing results according to the first embodiment. 6 is a flowchart illustrating details of a process for detecting a partition according to the first embodiment. 6 is a flowchart illustrating details at the start of a process for detecting a partition according to the first embodiment. 3 is a functional block diagram illustrating an outline of processing when the imaging unit according to the first embodiment acquires a wide-angle image. FIG. 6 is a functional block diagram illustrating processing by a distortion removing unit according to the first embodiment. FIG. FIG. 3 is an explanatory diagram of processing performed by a normal image conversion unit according to the first embodiment. It is explanatory drawing which shows the functional block of the process part of the space detection part of the present Example 2, and the process result of an image. It is a flowchart which shows the process of the space detection part of the present Example 2. It is explanatory drawing which shows the calculation process of the distance to the corner by the depth calculation part of the present Example 2. It is explanatory drawing which shows the process result of the human body detection part of a present Example 2, and the distance and direction of a corner.

  A present Example is described according to the following drawings.

  FIG. 1 is a front view illustrating an indoor unit 1, an outdoor unit 2, and a remote controller 3 of an air conditioner according to a first embodiment.

  The air conditioner shown in FIG. 1 includes an indoor unit 1, an outdoor unit 2, and a remote controller 3. The indoor unit 1 and the outdoor unit 2 are connected by refrigerant piping, and air-conditions the room in which the indoor unit 1 is installed by a known refrigerant cycle. Further, the indoor unit 1 and the outdoor unit 2 transmit and receive information to and from each other via a communication cable (not shown).

  The remote controller 3 is operated by a user and transmits an infrared signal toward a receiving unit included in the indoor unit 1. The indoor unit 1 has a receiving unit for receiving an infrared signal transmitted from the remote controller 3. The content of the infrared signal transmitted from the remote controller 3 is a command such as an operation request, a change in set temperature, a timer, a change in operation mode, or a stop request. Based on these signals, the air conditioner performs an air conditioning operation such as a cooling mode, a heating mode, or a dehumidifying mode.

  The indoor unit 1 according to the present embodiment includes a sensor 4, an electrical component 11, an imaging unit 26, and a temperature detection unit 27. The electrical component 11 has a processor and a memory. The electrical component 11 includes processing units such as the sensor 4, the detection unit 5, and the control unit 7.

  The processing unit included in the electrical component 11 may be a program or a physical device such as an integrated circuit.

  The sensor 4 is connected to the imaging unit 26 and the temperature detection unit 27, and holds image data captured by the imaging unit 26 and the temperature detection unit 27.

  The imaging unit 26 is, for example, a CCD (Charge Coupled Device) camera, and is a device that images a room. The imaging unit 26 is connected to an A / D converter and inputs an analog signal of captured image data to the A / D converter. The A / D converter is an electronic circuit that converts the image data input from the imaging unit 26 into a digital signal and outputs the digital signal. A digital signal of image data is input to the sensor 4 via the A / D converter. The imaging unit 26 may include an A / D converter, or the sensor 4 may include an A / D converter.

  The temperature detection unit 27 is, for example, a thermopile composed of 1 × 1 pixel, 4 × 4 pixel, or 1 × 8 pixel in the horizontal and vertical directions, and is a device that images the indoor temperature. The temperature detection unit 27 may be an infrared sensor or an infrared camera.

  FIG. 2 is an explanatory diagram illustrating a side sectional view of the indoor unit 1 according to the first embodiment.

  The indoor unit 1 includes the above-described sensor 4, electrical component 11, imaging unit 26, and temperature detection unit 27, as well as the housing base 8, the indoor fan motor 10, the vertical airflow direction plate 14, the front panel 15, and the air blowing. A mouth 19 is provided.

  The housing base 8 covers the outside of the indoor unit 1. The air outlet 19 is installed below the front surface of the indoor unit 1. The air whose temperature and humidity are adjusted by the indoor unit 1 is sent out of the indoor unit 1 from the air outlet 19.

  The indoor fan motor 10 is a device that takes in indoor air by rotating itself, and further sends air, the temperature or humidity of which has been adjusted by heat exchange, to the air outlet 19.

  The front panel 15 is installed so as to cover the front surface of the indoor unit 1. The front panel 15 may rotate according to a motor (not shown) for the front panel 15 with the lower end as an axis.

  The vertical wind direction plate 14 adjusts the wind direction of the air sent from the air outlet 19. The vertical wind direction plate 14 rotates according to an instruction from a microcomputer (not shown) provided in the indoor unit 1. The vertical wind direction plate 14 rotates according to a motor (not shown) for the vertical wind direction plate 14 with pivot shafts (not shown) provided at both ends as fulcrums.

  The imaging unit 26 and the temperature detection unit 27 illustrated in FIG. 2 are installed on the upper part of the air outlet 19. Then, the imaging unit 26 and the temperature detection unit 27 capture an image of the room from the gap between the up and down wind direction plates 14 and capture an image based on the horizontal line 37 and the optical axis 36. The optical axis 36 is a direction determined according to the directions of the lenses of the imaging unit 26 and the temperature detection unit 27, and is a direction in which the imaging unit 26 and the temperature detection unit 27 capture an image.

  Since the imaging unit 26 and the temperature detection unit 27 are installed on the front surface of the indoor unit 1, the imaging unit 26 and the temperature detection unit 27 transmit information about the room for controlling the adjustment of air by the indoor unit 1 to the indoor unit. 1 case base 8 etc. can acquire appropriately, without shielding.

  The indoor unit 1 accommodates internal structures such as a heat exchanger, left and right wind direction plates, an air suction port, a filter, and a blow-out air channel in addition to the above-described configuration. The indoor unit 1 may include a conventional sensor such as a humidity sensor, an illuminance sensor, an odor sensor, or an atmospheric pressure sensor.

  FIG. 3 is an explanatory diagram illustrating an imaging range by the imaging unit 26 according to the first embodiment.

  FIG. 3 shows a range in which the imaging unit 26 captures an image on a plane including the rotation direction of the imaging unit 26. FIG. 3 is an explanatory diagram illustrating a range in which the imaging unit 26 captures an image when the ceiling side is viewed from the floor side. When the indoor unit 1 is installed on the ceiling side of the room, the imaging unit 26 is installed with the lens facing downward so that the maximum angle between the optical axis 36 and the horizontal line 37 is a predetermined angle.

  The imaging unit 26 rotates around an axis inclined from the vertical direction by the same angle as the maximum angle between the horizontal line 37 and the optical axis 36. The imaging unit 26 images the room in parallel to the central axis (vertical in the image) and in the rotation direction (horizontal in the image).

  The viewing angle in the rotation direction of one imaging by the imaging unit 26 is, for example, approximately 60 °. In this case, the imaging unit 26 images the three directions of the left 32, the front 33, and the right 34 in the rotation direction. The imaging unit 26 is driven independently of other sensors (such as a humidity sensor and an odor sensor other than the sensor 4).

  The imaging unit 26 images each direction so that the viewing angles in the rotation direction overlap by a predetermined angle. For example, the 15 ° image captured on the left 32 and the 15 ° image captured on the front surface 33 are captured so as to overlap. The imaging unit 26 can appropriately arrange images captured in each direction in the order of imaging by associating overlapping image portions.

  The imaging unit 26 illustrated in FIG. 3 acquires an indoor image with a viewing angle of a total rotation direction of 150 ° by rotating 45 ° at a time. FIG. 3 shows viewing angles when an image of the entire room is divided by 15 ° in the rotation direction, and 10 viewing angles are shown.

  The imaging unit 26 shown below images the room in the order of left 32, front 33, and right 34, and then returns to the left 32 direction. However, the imaging unit 26 may image the room in any order as long as it can acquire an image of the entire room within a predetermined time, and may image the room in the order of the right 34, the front 33, and the left 32.

  FIG. 4 is an explanatory diagram illustrating a temperature detection range by the temperature detection unit 27 according to the first embodiment.

  FIG. 4 shows a range in which the temperature detection unit 27 detects the temperature when the ceiling side is viewed from the floor side. Similar to the imaging unit 26, the temperature detection unit 27 of the present embodiment is installed with the lens facing downward so that the maximum angle between the optical axis 36 and the horizontal line 37 is a predetermined angle. The difference in angle between the optical axis 36 of the temperature detection unit 27 and the optical axis 36 of the imaging unit 26 is set to almost zero so as to be set within a sufficiently small predetermined error.

  The temperature detection unit 27 rotates around an axis inclined from the vertical direction by the same angle as the maximum angle between the horizontal line 37 and the optical axis 36. The temperature detector 27 detects the temperature of the room in parallel with the central axis (vertical in the image) and in the rotation direction (horizontal in the image). The temperature detection unit 27 is driven independently of other sensors (such as a humidity sensor and an odor sensor other than the sensor 4).

  When the vertical detection range of the temperature detection unit 27 and the vertical imaging range of the imaging unit 26 are different, the temperature detection unit 27 and the imaging unit 26 may be installed so that the upper ends of the respective imaging ranges are aligned, It may be installed so that the vertical center position of each imaging range is aligned, or may be installed so that the lower ends of each imaging range are aligned.

  The overall viewing angle in the rotational direction of the temperature detection unit 27 and the overall viewing angle in the rotational direction of the imaging unit 26 are substantially the same. Further, the temperature detection unit 27 or the imaging unit 26 may be set to obtain a viewing angle substantially the same as the other viewing angle by changing the viewing angle so that one of the temperature detecting unit 27 or the imaging unit 26 rotates more than the other.

  The viewing angle of one detection by the temperature detection unit 27 is, for example, about 5 ° horizontal × 45 ° vertical when the lens of the temperature detection unit 27 captures an image with 1 horizontal pixel × 8 vertical pixels. Such a temperature detection unit 27 is rotated 30 times so that the detection ranges do not overlap from left to right, whereby the temperature detection unit 27 obtains a temperature detection result with a total viewing angle of 150 °.

  The temperature detection unit 27 according to the present embodiment images the room 30 times in order from left to right and then returns to the left. However, the temperature detection unit 27 of the present embodiment may detect the room temperature in any order as long as it can detect the temperature of the entire room within a predetermined time, and detects the temperature in the order of right, front, and left. Also good.

  In addition, as shown in FIG. 4, the temperature detection unit 27 shown below has a viewing angle of 5 ° in the rotation direction of one detection. However, the temperature detection unit 27 of the present embodiment may be the temperature detection unit 27 that can obtain a viewing angle of 150 ° or a sufficiently wide range by one imaging. When the temperature detection unit 27 can obtain a sufficiently wide range of viewing angles, the temperature detection unit 27 may not rotate.

  The drive timing of the imaging unit 26 is synchronized with the drive timing of the temperature detection unit 27. Specifically, the imaging unit 26 and the temperature detection unit 27 are driven at a timing at which the imaging range of the imaging unit 26 and the detection range of the temperature detection unit 27 overlap.

  For example, at the timing when the imaging unit 26 images the left 32, the temperature detection unit 27 starts detecting the temperature from the left, and at the timing when the imaging unit 26 images the front surface 33, the temperature detection unit 27 is the tenth time from the left. Start temperature detection.

  Thereby, the imaging unit 26 and the temperature detection unit 27 can capture or detect the indoor state in substantially the same time zone.

  As shown in FIG. 3, the imaging unit 26 of the present embodiment may have a viewing angle of 60 ° in the rotation direction of one imaging. On the other hand, the imaging unit 26 of the present embodiment may include a wide-angle lens that can capture a viewing angle of 150 ° or a sufficiently wide range with one imaging. When the imaging unit 26 includes a wide-angle lens, the imaging unit 26 may acquire an indoor image with only one imaging without rotating.

  Further, the imaging unit 26 may rotate 180 ° or more to image the periphery of the indoor unit 1.

  FIG. 5 is an explanatory diagram illustrating an imaging range by the imaging unit 26 including the wide-angle lens according to the first embodiment.

  The imaging unit 26 may be installed on the upper side surface or the lower side surface of the indoor unit 1 and may rotate 180 ° or more to image the periphery of the indoor unit 1. FIG. 5 shows a horizontal imaging range including the indoor unit 1.

  When the wide-angle lens included in the imaging unit 26 can capture a viewing angle of 150 °, the imaging unit 26 captures an image every time it rotates 120 °. Then, the imaging unit 26 images the ranges 350, 360, and 370 by rotating 240 degrees. Ranges 350, 360, and 370 each indicate an imaging range in the rotational direction in which imaging is performed once.

  The imaging unit 26 illustrated in FIG. 5 images each direction so that the viewing angles in the rotation direction overlap each other by 30 °. For example, the range 350 and the range 360 overlap the 30 ° range. As described above, the imaging unit 26 can appropriately arrange the images captured in each direction by associating the overlapping image portions.

  When the imaging unit 26 captures an imaging range of 360 °, the temperature detection unit 27 may also capture a temperature in the range of 360 °.

  FIG. 6 is an explanatory diagram illustrating a method of rotating the imaging unit 26 and the temperature detection unit 27 according to the first embodiment.

  The imaging unit 26 and the temperature detection unit 27 are rotated by a stepping motor 42. The stepping motor 42 is provided in the indoor unit 1 and is connected directly or indirectly to each of the imaging unit 26 and the temperature detection unit 27. Hereinafter, a plurality of methods by which the stepping motor 42 rotates the imaging unit 26 and the temperature detection unit 27 will be described.

  FIG. 6A is an explanatory diagram illustrating the imaging unit 26 and the temperature detection unit 27 that are rotated by the gear 43 according to the first embodiment.

  The indoor unit 1 may have a plurality of gears 43. And the stepping motor 42 shown to Fig.6 (a) may rotate the imaging part 26 or the temperature detection part 27 via the some gear 43 combined.

  FIG. 6B is an explanatory diagram illustrating the imaging unit 26 and the temperature detection unit 27 that are rotated by the arm 44 according to the first embodiment.

  The indoor unit 1 may have an arm 44. Then, the stepping motor 42 shown in FIG. 6B may rotate the imaging unit 26 or the temperature detection unit 27 via the arm 44.

  FIG. 6C is an explanatory diagram illustrating the imaging unit 26 and the temperature detection unit 27 that are directly rotated by the stepping motor 42 according to the first embodiment.

  The stepping motor 42 may be directly connected to the imaging unit 26 and the temperature detection unit 27, and may directly rotate the imaging unit 26 and the temperature detection unit 27.

  FIG. 7 is an explanatory diagram illustrating another example of positions where the imaging unit 26 and the temperature detection unit 27 according to the first embodiment are arranged.

  The imaging unit 26 and the temperature detection unit 27 illustrated in FIG. 1 are arranged in the horizontal direction. However, the imaging unit 26 and the temperature detection unit 27 may be arranged in the vertical direction as shown in FIG.

  Note that the imaging unit 26 and the temperature detection unit 27 may be arranged at any location in the indoor unit 1 as long as the room can be imaged or detected widely and are arranged close to each other. For example, when the indoor unit 1 is disposed in the lower part of the room, the imaging unit 26 and the temperature detection unit 27 may be disposed in the center of the front surface of the indoor unit 1 or in the upper part of the front surface.

  FIG. 8 is a functional block diagram illustrating an outline of a process for detecting the partition in the first embodiment and air-conditioning the room.

  The partition in the present embodiment is an object disposed between one room and another room or a hallway, and is, for example, a door, a bran, a shoji screen, or the like.

  The sensor 4 collects the image captured by the imaging unit 26 and the detection result detected by the temperature detection unit 27, and inputs the collected image and data such as the detection result to the detection unit 5.

  The electrical component 11 includes a detection unit 5, a control unit 7, and a storage unit 79. The storage unit 79 stores images acquired by the sensor 4 and provides data to the detection unit 5 and the control unit 7 as necessary. Further, the storage unit 79 stores the processing results of the detection unit 5 and the control unit 7 as necessary.

  The storage unit 79 may be implemented by a memory included in the electrical component 11 or may be implemented by a small and large-capacity storage device.

  The detection unit 5 includes at least a space detection unit 29. The detection unit 5 may include a human body detection unit 30 and a temperature matrix detection unit 31. The detection unit 5 detects the environmental features around the indoor unit 1 using the result obtained by the sensor 4.

  The space detection unit 29 according to the first embodiment holds an image picked up by the image pickup unit 26, and detects the presence or absence of a partition in the room in which the indoor unit 1 is installed based on the held image. Specifically, the space detection unit 29 extracts different regions between the past image and the current image, and analyzes the extracted region to determine whether there is a partition and whether the partition is open. Or closed state is detected. When it is detected that there is a partition, the space detection unit 29 detects the position, area, shape, complexity, brightness, and the like of the partition.

  The space detection unit 29 detects and removes distortion from the captured image when the imaging unit 26 captures an image with a wide-angle lens having a predetermined viewing angle, such as 150 °.

  The human body detection unit 30 uses the image captured by the imaging unit 26 to detect the presence / absence of an occupant and information related to the occupant. The human body detection unit 30 detects, for example, the position, the number of persons, the amount of activity, the attribute, the individual, the life scene, and the like of the occupant as information on the occupant.

  The temperature matrix detection unit 31 detects the surface temperature of the room where the indoor unit 1 is installed using the detection result of the temperature detection unit 27. The temperature matrix detection unit 31 detects a temperature matrix, for example.

  The processing units of the detection unit 5 such as the space detection unit 29, the human body detection unit 30, and the temperature matrix detection unit 31 notify the control unit 7 of the detected result as information. The control unit 7 controls the adjustment of air by the air conditioner based on the information notified from the detection unit 5. Further, the control unit 7 may control the adjustment of air by the air conditioner based on data input from a humidity sensor other than the sensor 4 or data input from the remote controller 3.

  The control unit 7 is connected to, for example, the indoor fan motor 10 or the compressor, the outdoor fan motor, the four-way valve, or the motor operated valve provided in the outdoor unit 2, and controls the operation of these devices to control the air conditioner. To control. Specifically, the control unit 7 controls the air conditioner to adjust the air volume to the room, the wind direction, the temperature of the air sent to the room, and the like.

  The four-way valve is a valve provided in the outdoor unit 2 for switching the operation of the air conditioner to the cooling mode or the heating mode. When the four-way valve is switched, the air flow is reversed, and for example, the roles of the outdoor unit 2 that has cooled the high-temperature and high-pressure air and the indoor unit 1 that has taken away the heat in the room are reversed.

  The electric valve is a valve provided in the outdoor unit 2 for expanding the refrigerant. The outdoor fan motor is a motor that is provided in the outdoor unit 2 and takes in outside air.

  FIG. 9 is a flowchart illustrating processing for detecting a partition when the imaging unit 26 of the first embodiment rotates.

  The process illustrated in FIG. 9 illustrates a process in which the space detection unit 29 detects a partition when the imaging unit 26 acquires an image of the entire room by imaging three times. The imaging unit 26 may start the processing of FIG. 9 in a predetermined cycle set in advance, or when it is determined in step 77 that the partition cannot be detected, or in step 78 information on the partition is sent to the control unit 7. After being notified, the processing of FIG. 9 may be started (70).

  After step 70, the imaging unit 26 captures an image in the first direction. The image in the first direction is, for example, the image in the left 32 direction shown in FIG. 3 and the image in the range 350 direction shown in FIG. And the imaging part 26 sends the imaged image to the sensor 4 via an A / C converter (71).

  In step 71, the temperature detector 27 detects the temperature in the direction corresponding to the first direction. For example, the temperature detection unit 27 illustrated in FIG. 4 detects the indoor temperature at least nine times in order from the left. Moreover, when the indoor unit 1 has the imaging unit 26 shown in FIG. 5, the indoor temperature is detected at least five times while rotating. Then, the temperature detection unit 27 sends the detected result to the sensor 4.

  When the detection unit 5 acquires an image captured by the imaging unit 26 via the sensor 4, the space detection unit 29 analyzes the captured image, and when the image in the first direction includes a partition, A partition is detected (72). In step 72, the human body detection unit 30 detects the presence or absence of a person in the first direction based on the detection result by the temperature detection unit 27, and the temperature matrix detection unit 31 detects the detection result by the temperature detection unit 27. Based on the above, a temperature matrix in the first direction may be output.

  Step 73 and step 74 are the same as the processing of step 71 and step 72, and are processing for capturing and analyzing an image in the second direction. The image in the second direction is, for example, the image in the left 32 direction shown in FIG. 3 and the image in the range 360 direction shown in FIG.

  Further, Step 75 and Step 76 are the same as the processing of Step 71 and Step 72, and are processing for capturing and analyzing an image in the third direction. The image in the third direction is, for example, the image in the left 32 direction shown in FIG. 3 and the image in the range 370 direction shown in FIG.

  Step 73 is executed after step 71, and step 75 is executed after step 73. Step 74 is executed when the detection unit 5 receives data such as an image after step 72. Step 76 is executed when the detection unit 5 receives data such as an image after Step 75.

  After step 76, the space detector 29 determines whether a partition is detected in steps 72, 74, and 76 (77). If no partition is detected, the process shown in FIG. When the partition is detected, the detection unit 5 notifies the control unit 7 of the partition state and the partition position information (78).

  The partition state information in this embodiment is information indicating whether the partition is in an open state (open state) or whether the partition is in a closed state (closed state). Further, the partition position information indicates at least one of the center coordinates of the partition, the coordinates of the outer periphery of the partition (for example, indicated by the coordinates of two points on the upper right and the lower left of the partition), and the direction of the partition. .

  Note that the detection unit 5 in step 78 may notify the control unit 7 of the surface temperature of the partition and the surface temperature of the region other than the partition based on the processing result of the temperature matrix detection unit 31.

  After step 78, the process shown in FIG. 9 returns to step 70.

  When the imaging unit 26 acquires an image of the entire room by imaging twice or four times or more, processing corresponding to step 71 and step 72 is executed for each imaging direction. Further, when the imaging unit 26 acquires an image of the entire room by a single imaging, the processes corresponding to step 71 and step 72 are each executed only once.

  The control unit 7 controls the adjustment of the air by the air conditioner based on the partition state and the partition position information notified from the detection unit 5. For example, when the partition is in the open state, the control unit 7 causes each device of the air conditioner to adjust the air flow so as to increase the amount of air sent in the direction of the partition.

  Further, when the information notified from the detection unit 5 indicates that there is a difference of a predetermined temperature or more between the surface temperature of the partition and the surface temperature of the region other than the partition, the control unit 7 The air volume and direction directed to the partition may be adjusted so that there is no more.

  Thereby, even when the partition is in the open state, the air conditioner of the first embodiment can adjust the air in the vicinity of the partition and can adjust the entire room to a comfortable state.

  FIG. 10 is an explanatory diagram illustrating the functional blocks of the processing unit of the space detection unit 29 and the image processing result according to the first embodiment.

  The space detection unit 29 according to the first embodiment includes at least processing units of a distortion removal unit 290, a difference identification unit 291, an area division unit 292, a complexity calculation unit 294, a partition position detection unit 295, and a partition opening / closing determination unit 296. . The space detection unit 29 includes a partition determination unit 293 as necessary.

  The processing unit included in the space detection unit 29 may be implemented by a program or a physical integrated circuit.

  Image A and image B shown in FIG. 10 include a partition on the right. An image A shown in FIG. 10 includes an open partition, and an image B illustrated in FIG. 10 includes a closed partition. Image A is an image captured after image B is captured.

  When the imaging unit 26 has a wide-angle lens, in the process illustrated in FIG. 10, the distortion removing unit 290 first removes the distortion of the image A and the image B. Here, when the removed flag indicating that the distortion has already been removed is added to the image A or the image B, the distortion removing unit 290 does not remove the distortion from the image to which the removed flag is added.

  A method of removing distortion by the distortion removing unit 290 will be described later. By using the image A and the image B captured by the wide-angle lens, the control unit 7 can detect a door or a beam that is not reflected in the image captured in a narrow range.

  The difference specifying unit 291 compares two input images (image A and image B), and specifies a different region (difference region 190) between the image A and the image B. The image 90 shows the identified difference area 190.

  The area dividing unit 292 divides the difference area 190 specified by the difference specifying part 291 from other areas as a candidate area 191 of an area including a partition. The image 91 shows the divided candidate area 191.

  The area dividing unit 292 specifies the position of the specified difference area 190 in the image A. Thereby, the coordinates (such as the coordinates of the outer periphery of the partition) in the image A of the partition or the direction of the partition are determined.

  The space detection unit 29 executes a partition determination unit 293 as necessary, and the partition determination unit 293 determines whether the candidate region 191 is a region including a partition. When the partition determination unit 293 is not executed, the space detection unit 29 determines that the candidate region 191 is a region including the partition.

  The complexity calculation unit 294 calculates the complexity of the area including the partition in the image A. The partition position detection unit 295 detects the center position 93 of the region including the partition. The image 92 shows the detected center position 93.

  The partition opening / closing determination unit 296 determines whether the partition in the image A is in an open state or a closed state based on the complexity calculated by the complexity calculation unit 294.

  FIG. 11 is a flowchart illustrating details of processing for detecting a partition according to the first embodiment.

  More specifically, FIG. 11 is a flowchart expressing the processing flow of the processing unit shown in FIG. 10 in detail.

  The process shown in FIG. 11 shows the details of the process in which the space detection unit 29 detects the partition, and corresponds to the processes in steps 72, 74, and 76 shown in FIG. Moreover, the process shown in FIG. 11 shows a process in case the partition determination part 293 is performed.

  When the space detection unit 29 receives the image data (image A) captured by the imaging unit 26 via the sensor 4, the space detection unit 29 reads the previously received image data (image B). Specifically, the space detection unit 29 holds the received image data in the memory for each direction in which the image is captured, and when the image A is received from the sensor 4, the space detection unit 29 has the same direction as the direction in which the image A was captured. Image B is read from memory.

  Note that the space detection unit 29 may store information related to the partition (partition state and position, etc.), which is a result of the processing of FIG. 11 executed in the past, and image data in association with each other in the memory. Good. Then, when the image B is read, information related to the partition corresponding to the image B may also be read.

  After the start of the process shown in FIG. 11, the distortion removing unit 290 determines whether distortion has been removed from both the image A and the image B (52). The distortion removing unit 290 determines that the distortion is removed from the image when the removed flag is added to the image.

  When the distortion has been removed from both the image A and the image B, the difference specifying unit 291 executes Step 54.

  When at least one of the image A and the image B has not been subjected to distortion removal, the distortion removal unit 290 specifies a region from which distortion is to be removed in the image A and the image B to which the distortion removal flag is not added. Further, the image is converted so as to remove the distortion from the specified region (53). Then, a removed flag is added to the image from which distortion has been removed.

  Here, the distortion removing unit 290 may specify a predetermined range of the image A or the image B, for example, an area having a predetermined length or more from the center point as an area for removing distortion. This is because a region where distortion is generally generated is a region on the outer side of the imaging range.

  In addition, the distortion removing unit 290 may specify all areas in the image as areas for removing distortion when the processing amount is not limited. A method for removing the distortion will be described later. The distortion removing unit 290 inputs the image A and the image B from which the distortion is removed to the difference specifying unit 291.

  When the distortion removing unit 290 removes distortion from the image, an accurate image can be acquired even when the image is captured using the wide-angle lens.

  If it is determined in step 52 that both the image A and the image B have the distortion removed, or after step 53, the difference specifying unit 291 executes step 54. The difference specifying unit 291 specifies the difference area 190 between the image A and the image B, and specifies the partition area (54). The images A and B used here are images from which distortion has been removed in step 53 or the previous processing.

  The difference specifying unit 291 specifies the difference area 190 using a color difference between the image A and the image B, a difference in luminance, or the like.

  If the partition is included in the image B and the partition is opened or closed between the time when the image B is captured and the time when the image A is captured, the space between the image A and the image B , Regions having different images, that is, a difference region 190 is included.

  After step 55, the space detection unit 29 determines whether or not at least one difference region 190 has been specified in step 55 (56). When no difference area 190 is specified, the space detection unit 29 ends the process shown in FIG. 11 (59), and ends the process of step 72, 74, or 76.

  When at least one difference area 190 is specified in step 55, the area dividing unit 292 divides the difference area 190 from the image A as a candidate area 191. Then, the position (peripheral coordinates, direction, etc.) in the image A of the candidate area 191 is specified (57). In step 57, by dividing the difference area 190 as the candidate area 191 from the image A, the detection unit 5 can detect an area including a partition from the difference area 190.

  After step 57, the space detection unit 29 determines whether or not the processing from step 60 has been performed on all candidate areas 191 (58). If there is one candidate area 191, step 58 is not necessary. When the process of step 60 is performed on all candidate areas 191, the space detection unit 29 ends the process shown in FIG. 11 (59), and ends the process of step 72, 74, or 76.

  When there is an area in the candidate area 191 where the process from step 60 has not been executed, the space detection unit 29 selects a candidate area 191 that has not executed the process in step 60 (hereinafter, candidate area C). And the partition determination part 293 calculates the circularity of the candidate area | region C (60). In step 60, the partition determination unit 293 specifies the shape of the candidate area C.

  The circularity in the present embodiment is an evaluation value with which a higher value is calculated as the shape of the region is closer to a perfect circle.

  After step 60, the partition determination unit 293 determines whether the candidate region C is close to a circle by determining whether the circularity calculated in step 60 exceeds a predetermined threshold (61). .

  When the calculated circularity exceeds a predetermined threshold and it is determined that the candidate area C is closer to a circle than the predetermined reference, it is unlikely that the candidate area C is a partition in this embodiment. For this reason, the partition determination unit 293 excludes the candidate region C from the candidate region 191 of the region including the partition. Then, the processing shown in FIG.

  When the calculated circularity is less than or equal to a predetermined threshold and it is determined that the candidate area C is not circular compared to a predetermined reference, it is highly likely that the candidate area C is an area including a partition in this embodiment. . For this reason, the partition determination unit 293 calculates the area of the candidate region C and specifies the size of the partition (62).

  In step 61, the partition determination unit 293 uses the predetermined minimum value and the predetermined maximum value larger than the predetermined minimum value to determine whether the circularity of the candidate area C is an appropriate circularity as a partition. It may be determined. Specifically, the partition determination unit 293 may exclude the candidate region C from the region candidates including the partition even when the circularity of the candidate region C is equal to or less than a predetermined minimum value. This is because when the circularity of the candidate area C is less than or equal to a predetermined minimum value, the candidate area C is highly likely to be an area having a complicated shape.

  After step 62, the partition determination unit 293 determines whether or not the area calculated in step 62 is equal to or smaller than a predetermined minimum value (63). When the area of the candidate area C is equal to or less than the predetermined minimum value, the candidate area C is highly likely to be a laundry or a human body, and is unlikely to be an area including a partition. For this reason, the partition determination unit 293 determines that the area of such a candidate region C is not appropriate as the partition size, and excludes the candidate region C from the candidate region 191 of the region indicating the partition. Then, the processing shown in FIG.

  If it is determined in step 63 that the area of the candidate region C exceeds the predetermined minimum value, the candidate region C is determined to be a region including a partition, and step 64 is executed. By executing Steps 61 and 63, the partition determination unit 293 accurately detects a region that is more likely to be a region including a partition from the candidate regions 191, and the detected region includes a partition. Can be determined as

  In step 63, the partition determination unit 293 uses the predetermined minimum value and the predetermined maximum value sufficiently larger than the predetermined minimum value to determine whether the area of the candidate region C is an appropriate area for the partition. It may be determined. Specifically, the partition determination unit 293 may exclude the candidate region C from the region candidates including the partition even when the area of the candidate region C is equal to or larger than a predetermined maximum value. This is because the candidate area C is an area of a difference caused by a change in indoor illumination intensity or a large change in room illuminance or color when the area of the candidate area C is equal to or greater than a predetermined maximum value. This is because there is a high possibility.

  The complexity calculation unit 294 calculates the complexity in the image A of the area determined to be the area including the partition in step 61 and step 63 (64). The complexity calculation unit 294 calculates the complexity based on the number of edges included in the determined area.

  The complexity in this embodiment is the density of objects (walls, floors, furniture, people, etc.) included in an image, and is an amount such as color or luminance included in the image. Further, the higher the complexity of the present embodiment, the higher the number of edges included in the region.

  An image having a predetermined size on which a distant landscape is displayed generally includes more objects than the region having the predetermined size including a nearby landscape. For this reason, when the partition is closed, only one partition is displayed in the area including the partition. However, when the partition is open, the partition area includes many objects. When the complexity calculation unit 294 calculates the complexity, the partition opening / closing determination unit 296 can perform processing described later.

  After step 64, the partition position detector 295 calculates the coordinates of the center position (corresponding to the center position 93 shown in FIG. 10) of the region determined to be the region including the partition (65). This is because the control unit 7 adjusts the wind direction by notifying the control unit 7 of the coordinates of the center position calculated by the space detection unit 29. The partition position detection unit 295 may calculate the coordinates of the center position using any method as long as the indoor unit 1 can determine the wind direction.

  After step 65, the partition opening / closing determination unit 296 determines whether or not the calculated complexity in the image A is equal to or less than a predetermined threshold (66). Thereby, the partition opening / closing determination unit 296 determines whether the partition included in the determined region is in an open state or a closed state.

  In general, an area including an open partition has a higher degree of complexity than an image of an area including a closed partition because it includes a person, laundry, furniture, or the like existing in an adjacent room or hallway. For this reason, the partition opening / closing determination unit 296 determines in step 66 that the partition indicated by the determined region is in an open state when the complexity calculated in step 64 is higher than a predetermined threshold. Then, the partition opening / closing determination unit 296 outputs the position of the area specified in step 57 (peripheral coordinates, direction, etc.), the center position (coordinates) calculated in step 64, and the fact that the partition is opened ( Output 68).

  In step 66, when the complexity calculated in step 64 is equal to or less than a predetermined threshold, the partition opening / closing determination unit 296 determines that the partition indicated by the determined area is in a closed state. Then, the position of the area specified in step 57 (peripheral coordinates, direction, etc.), the center position (coordinates) calculated in step 64, and the fact that the partition is closed are output as processing results (output). 67).

  In step 66, when the complexity of the determined area in the image B is lower than the complexity of the determined area in the image A, the partition opening / closing determination unit 296 opens the partition indicated by the determined area. You may determine that it is in a state.

  After the output 67 or the output 68, the process shown in FIG. 11 returns to step 58, and the space detection unit 29 selects a new candidate area C from the candidate area 191.

  When it is determined in step 77 shown in FIG. 9 that the output 67 or the output 68 is output as a result of the processing of FIG. 11 performed on the image in any direction, the space detection unit 29 detects the partition. It is determined that Then, the space detection unit 29 notifies the control unit 7 of the output 67 or the output 68.

  The partition determination unit 293 accurately determines whether or not the candidate region C is a region including a partition by performing the following processing after step 58 and before step 64 is started. May be.

  In step 61 described above, it is determined whether or not the candidate region C is a region including a partition according to the circularity of the candidate region C. However, the partition determination unit 293 may instead of the circularity or in addition to the circularity. Depending on the straightness of the upper end, lower end, left end, or right end of the candidate area C, it may be determined whether the candidate area C is an area including a partition. This is because a partition such as a door or a bran is generally rectangular in many cases.

  For example, when the calculated linearity indicates that at least one end is not a straight line, the candidate area C includes a curtain, a figurine, or a laundry. Therefore, the partition determination unit 293 includes a candidate area 191 that includes a partition. The candidate area C may be excluded from

  Furthermore, the partition determination unit 293 may determine whether the partition is a partition based on the position of the upper end of the candidate area C in the image A. Specifically, when the position of the upper end of the candidate area C is a predetermined distance or more from the upper end of the image A, the candidate area C is highly likely to be a television set at a certain height below the ceiling. Therefore, the partition determination unit 293 may exclude the candidate region C from the candidate region 191 of the region including the partition.

  Furthermore, when the difference between the upper end angle of the candidate area C and the upper end angle of the image A is not within a predetermined angle range, the partition determination unit 293 selects the candidate area C from the candidate area 191 of the area including the partition. It may be excluded.

  Further, the partition determination unit 293 may determine whether the region includes the partition based on the position of the left end or the right end of the candidate region C. Specifically, the partition determination unit 293 partitions the candidate region C into partitions when the difference between the left end or right end angle of the candidate region C and the left end or right end angle of the image A is not within a predetermined range. May be excluded from the candidate area 191 of the area including.

  Further, the partition determination unit 293 may determine whether the region includes a partition based on the luminance of the candidate region C. Specifically, when the brightness of the candidate area C between the image A and the image B changes more than a predetermined range, the candidate area C is a difference caused by a change in brightness due to opening / closing of the curtain or lighting of the illumination. The partition determination unit 293 may exclude the candidate region C from the candidate region 191 of the region including the partition.

  Furthermore, the partition determination unit 293 may determine whether or not the candidate region C is a region including a partition based on the complexity calculated a plurality of times in the past. For example, when the complexity of the candidate area C in three or more images captured in the past changes three or more times to different values, the candidate area C is likely to be an image including a television. For this reason, the partition determination unit 293 may exclude such a candidate region C from the candidate regions 191 of the region including the partition.

  This is because the image displayed on the television always changes, and the complexity calculated for the television image changes variously. On the other hand, the complexity calculated in the region including the partition is considered to change in two patterns. As described above, the partition determination unit 293 can exclude the region including other than the partition from the candidate region 191 using the complexity, and can improve the detection accuracy of the partition.

  Furthermore, when the temperature matrix detection unit 31 measures the indoor surface temperature, the space detection unit 29 may detect the partition with high accuracy by using the measurement result of the temperature matrix detection unit 31. For example, the area dividing unit 292 may divide the area where the surface temperature is different between the image A and the image B from the image A as the candidate area 191 in step 57.

  Further, based on the temperature distribution of the surface temperature measured by the temperature matrix detection unit 31, the region division unit 292 and the partition determination unit 293 may calculate the position and area of the partition.

  Further, when the surface temperature of the candidate area C is higher than the predetermined maximum temperature or lower than the predetermined minimum temperature, the candidate area C is likely to include furniture such as an electrical appliance or an aquarium. The candidate area C may be excluded from the candidate area 191 of the area including the partition.

  Further, as a result of measurement by the temperature matrix detection unit 31, a difference between the room temperature and the surface temperature of the candidate region C, or a difference between the surface temperature around the candidate region C and the surface temperature of the candidate region C is a predetermined value. If it is smaller than the difference, the candidate area C is likely to be a laundry or the like. For this reason, the partition determination unit 293 may exclude such a candidate region C from the candidate regions 191 of the region including the partition.

  Further, when the information related to the partition specified in the image B or a plurality of past images has been read in step 55, the partition determination unit 293 determines that the candidate region C performs partitioning based on the read information. You may determine whether it is an area | region to include. For example, when the position, area, shape, and the like of a region that has been determined to be a region including a partition in the past and the position, area, shape, etc. of the candidate region C are approximated according to a predetermined criterion, the partition determination unit 293 may determine that the candidate area C is an area including a partition.

  In addition, when the partition determination unit 293 obtains predetermined certainty that the candidate area C is an area including the partition based on the read information regarding the past partition, the candidate area C includes the partition. The region may be determined.

  Moreover, the partition determination part 293 is the area | region determined as the area | region which included the television or the laundry in the past, when the information regarding the area | region including a television or laundry is included in the read information regarding the past partition And the candidate area C may be the same or approximate, the candidate area C may be excluded from the candidate area 191 of the area including the partition.

  Below, the partition opening / closing determination part 296 shows the method of the process which determines the opening / closing of a partition other than the above-mentioned process.

  In step 66, the partition opening / closing determination unit 296 may determine whether the partition is in an open state or a closed state based on the measurement result of the temperature matrix detection unit 31.

  For example, when the surface temperature of the region including the partition is approximately the same as that around the region, the partition opening / closing determination unit 296 may determine in step 66 that the partition is in the closed state. Thereby, even when the temperature of the adjacent room through the partition and the temperature of the room in which the indoor unit 1 is installed become uniform, the partition opening / closing determination unit 296 performs air conditioning toward the partition other than the partition. The controller 7 can be instructed appropriately so that the air conditioning is in the same direction as the direction.

  Further, for example, the temperature matrix detection unit 31 measures the indoor air temperature and the outdoor air temperature. And the temperature matrix detection part 31 approximates the room temperature when a partition is opened, and the room temperature when a partition is closed by learning the room temperature and outdoor temperature measured several times. Then, the space detection unit 29 divides the space between the time when the image B is captured and the time when the image A is captured based on the room temperature when the image B is captured, the room temperature when the image A is captured, and the estimated room temperature. It may be determined whether is open or closed.

  The indoor unit 1 described above is installed on the ceiling side of the room, and the imaging unit 26 is installed in the lower part of the indoor unit 1. However, the imaging unit 26 may be installed in any direction as long as it can image a room widely. For example, when the indoor unit 1 is installed close to the floor, the lens of the imaging unit 26 faces upward. It may be installed. In this case, it is desirable that the imaging unit 26 be installed so that the entire partition can be imaged.

  In step 53 shown in FIG. 11, the distortion removing unit 290 specifies an area from which distortion is removed based on the position of the image. On the other hand, the distortion removing unit 290 may specify the difference area between the image B and the image A specified by the difference specifying unit 291 as an area for removing distortion. FIG. 12 shows processing A (steps 52 to 54 in FIG. 11) executed in this case.

  FIG. 12 is a flowchart illustrating details at the start of the process of detecting the partition according to the first embodiment.

  Process A shown in FIG. 12 includes steps 51 to 53.

  First, in the process of FIG. 12, the difference specifying unit 291 specifies the difference area 190 between the image A and the image B, and specifies the partition area (51). The specific contents are the same as in step 54 shown in FIG. Then, information such as the position of the difference area 190 is stored in the memory.

  Note that, in step 51, when one of the image A and the image B has the distortion removed and the other has the distortion removed, the difference specifying unit 291 acquires an image from which the distortion has not been removed from the memory, A difference area 190 between the image B from which distortion has not been removed and the image A from which distortion has not been removed is specified.

  After step 51, the distortion removing unit 290 determines whether distortion has been removed from both the image A and the image B (52). The distortion removing unit 290 determines that the distortion is removed from the image when the removed flag is added to the image.

  When the distortion has been removed from both the image A and the image B, the space detection unit 29 executes step 56. And the process after step 56 shown in FIG. 11 is performed.

  If at least one of the image A and the image B has not been subjected to distortion removal, the distortion removal unit 290 identifies the difference area 190 in the image A as an area from which distortion is to be removed, and removes distortion from the difference area 190 (53). . Then, a removed flag indicating that the distortion has been removed from the difference area 190 is added to the image A from which the distortion has been removed.

  As a result, useless processing such as removing distortion from the area other than the difference area 190 can be omitted.

  After step 53, the space detection unit 29 executes the processing after step 56.

  In the process A shown in FIG. 12, by removing the distortion only from the difference area 190 specified by the difference specifying section 291, the distortion removing section 290 removes the distortion from all the images or removes the distortion from a wide area. The amount of processing can be reduced.

  FIG. 13 is a functional block diagram illustrating an outline of processing performed when the imaging unit 26 according to the first embodiment acquires a wide-angle image.

  The wide-angle imaging unit 80 of Example 1 is the imaging unit 26 that includes a wide-angle lens. The imaging unit 26 stores the captured image in the storage unit 79 of the electrical component 11.

  Then, the conversion unit 2901 included in the detection unit 5 identifies an area from which the distortion of the image is to be removed, and converts the image so as to remove the distortion of the identified area. The conversion unit 2901 is a distortion removal unit 290 included in the space detection unit 29.

  Then, the space detection unit 29, the human body detection unit 30, the temperature matrix detection unit 31 and the like in the detection unit 5 extract indoor features from the image and store information indicating the indoor features in the storage unit 79. Here, the indoor features are the partition detected by the space detection unit 29 and the opening / closing of the partition.

  And the control part 7 adjusts the wind direction and air volume of the wind sent indoors based on the information which shows the indoor feature stored in the memory | storage part 79. FIG.

  FIG. 14 is a functional block diagram illustrating processing performed by the distortion removing unit 290 according to the first embodiment.

  The distortion removing unit 290 includes a distortion removing region specifying unit 110, a normal image converting unit 111, and an interpolation processing unit 112. The distortion removing unit 290 performs image conversion on a region from which distortion is removed in the image by the processing of FIG.

  First, the distortion removal area specifying unit 110 specifies an area from which distortion is to be removed. Specifically, the distortion removal region specifying unit 110 may specify a predesignated range in the image as a region for removing distortion, or a difference region from the past image specified by the difference specifying unit 291. May be specified as a region from which distortion is removed.

  The normal image conversion unit 111 uses the reference point (corresponding point) in the (x, y) plane, which is the region before distortion removal, specified as the region from which distortion is removed, as the region after distortion removal (u, v ) Corresponding to points on the plane (planarized image), the corresponding points in the planarized image are determined.

  FIG. 15 is an explanatory diagram of normal image conversion processing by the normal image conversion unit 111 according to the first embodiment.

The normal image conversion unit 111 uses the relationship between the (u, v) plane and the (x, y) plane as shown in FIG. Correspondence with corresponding points on the (x, y) plane is performed to determine corresponding points on the (u, v) plane. Accordingly, the normal image conversion unit 111 converts the corresponding point on the (x, y) plane to the corresponding point on the (u, v) plane.









α = rotation angle β = zenith angle φ = rotation angle of the extracted plane m = magnification R = radius of the outer circumference circle of the wide-angle image

  After the processing by the normal image conversion unit 111, the interpolation processing unit 112 generates a pixel on the (u, v) plane by interpolating from around the corresponding point on the (u, v) plane, for example, using cubic interpolation. To do.

  The distortion removing unit 290 removes distortion from the area specified as the area from which distortion is removed by the processing shown in FIG. 14, and as a result, converts the input image into an image from which distortion has been removed. In subsequent processing, an image from which distortion has been removed is used, so that indoor features (such as partitions) can be accurately detected.

  According to the first embodiment, the air conditioner according to the first embodiment is a space in which the air cannot be adjusted in the vicinity of the room partition in order to appropriately detect the state and position of the room partition and adjust the air volume and the wind direction according to the detection result. It is possible to comfortably adjust the air in the entire room without generating any.

  Moreover, the air conditioner of Example 1 specifies the area | region which removes distortion from the image imaged by the imaging part 26, and also removes distortion from the specified area | region. For this reason, even when the imaging unit 26 captures a wide-width image using a wide-angle lens, distortion can be removed from the image, and indoor features can be accurately detected. Further, for example, by capturing a partition that is not captured by a camera that captures a narrow range, and detecting the opening and closing of the partition, it is possible to prevent the accumulation of heat.

  The detection unit 5 according to the second embodiment detects a floor plan using the image captured by the imaging unit 26. The indoor floor plan in the present embodiment means at least the shape and size of the room indicated by the distance from the corner of the room (hereinafter, corner) to the indoor unit 1 and the angle from the front direction of the indoor unit 1 to the corner. Say. In addition, the depth of the room in the present embodiment is a distance from the indoor unit 1 to the farthest corner.

  The indoor unit 1, the outdoor unit 2, and the remote controller 3 of the second embodiment are the same as those of the first embodiment. Further, the sensor 4, the imaging unit 26, and the temperature detection unit 27 of the second embodiment are the same as those of the first embodiment. The electrical component 11 according to the second embodiment is the same as the electrical component 11 according to the first embodiment except for the space detection unit 29.

  The space detection unit 29 according to the second exemplary embodiment is a processing unit that detects a floor plan using an image captured by the imaging unit 26. The detection unit 5 according to the second embodiment may be implemented by a program or a physical device such as an integrated circuit.

  Below, the case where the indoor unit 1 is installed in the ceiling side of a room is mainly shown.

  FIG. 16 is an explanatory diagram illustrating functional blocks of the processing unit of the space detection unit 29 according to the second embodiment and an image processing result.

  The space detection unit 29 according to the second embodiment includes processing units such as a distortion removal unit 290, an edge detection unit 81, a straight line generation unit 82, an intersection detection unit 83, and a depth calculation unit 84. The processing unit of the space detection unit 29 according to the second embodiment may be implemented by a program or a physical device. The space detection unit 29 of the second embodiment may include the processing unit of the second embodiment and the same processing unit as the space detection unit 29 of the first embodiment, and may detect a floor plan and a partition. .

  When the image 94 shown in FIG. 16 is input via the sensor 4, the distortion removing unit 290 converts the input image 94 into an image 94 from which distortion is removed by the same method as the process shown in FIG. 14 of the first embodiment. Convert.

  After the processing by the distortion removing unit 290, the edge detecting unit 81 detects an edge included in the image 94 from which the distortion is removed. An image 95 shown in FIG. 16 is a result of detecting an edge from the image 94.

  The straight line generation unit 82 extends a selected edge among the edges detected by the edge detection unit 81. An image 96 shown in FIG. 16 shows a straight line generated as a result of the edge being extended by the straight line generation unit 82.

  The intersection detection unit 83 detects the intersection of the straight lines extended by the straight line generation unit 82. The image 97 shows a plurality of intersection points 98 (98a to 98c) detected in the image 96.

  The depth calculation unit 84 calculates the distance to the corner according to the position of the intersection 98 in the image 97, and obtains the direction toward the corner. Then, the depth calculation unit 84 specifies the farthest corner from the indoor unit 1 in the room based on the calculated corner distance, and determines the distance to the farthest corner as the depth of the room.

  FIG. 17 is a flowchart illustrating the processing of the space detection unit 29 according to the second embodiment.

  When an image captured by the imaging unit 26 is input, the process illustrated in FIG. 17 is started. In addition, the space detection unit 29 of the second embodiment is similar to the processing of the space detection unit 29 of the first embodiment illustrated in FIG. 9 every time an image captured in each direction by the imaging unit 26 is input to the detection unit 5. You may perform the process shown in FIG.

  Further, the space detection unit 29 according to the second embodiment determines whether or not a corner is detected by the process illustrated in FIG. 17 in the same manner as Step 77 illustrated in FIG. 9. Notify And the control part 7 controls the quantity or direction of the wind which an air conditioner sends according to the information regarding the notified floor plan.

  First, the distortion removing unit 290 converts the input image into an image from which distortion has been removed by the same method as the process shown in FIG. 14 of the first embodiment (85). By detecting the floor plan using the image from which the distortion is removed by the distortion removing unit 290, for example, when the imaging unit 26 having a wide-angle lens captures a wide-angle image, a beam or the like that is not reflected in an image with a narrow viewing angle Can be accurately detected.

  Then, the edge detection unit 81 detects an edge included in the input image (86). An edge is a boundary line between two different colors that appear in an image, and in particular, in the present embodiment, is a straight line.

  After step 86, the straight line generation unit 82 extracts an edge having a predetermined length from the edge detected in step 86, and generates a straight line by extending the extracted edge in the image (87).

  After step 87, the intersection detection unit 83 detects the intersection 98 of the straight line generated in step 87 (88). The intersection 98 detected here includes an intersection corresponding to a corner in the room where the indoor unit 1 is installed. If the intersection point 98 is not detected in step 88, the process shown in FIG. 17 ends.

  Note that when three or more straight lines do not intersect at one intersection and there are a plurality of intersections of two or more straight lines within a predetermined position range, the intersection detection unit 83 averages the positions of the plurality of intersections. The average position may be detected as the position of the intersection 98.

  After step 88, the depth calculation unit 84 extracts the intersection point 98 (the vanishing point in this embodiment) corresponding to the corner from the intersection point 98 detected in step 88 according to a predetermined criterion. The predetermined reference is, for example, when the imaging unit 26 captures an image below the horizontal plane, is located at the lowest position in the image, or is located on the same straight line (excluding the vertical direction) as the lowest intersection point 98 It is to be.

  This is because, for example, when a plurality of intersections 98 are detected as in an image 97 shown in FIG. 16, an intersection 98c corresponding to a corner where the floor and two walls intersect is detected as a vanishing point. In addition, the vanishing point in a present Example shows the point corresponding to a corner in an image.

  Then, the depth calculation unit 84 calculates the distance from the indoor unit 1 to the corner based on the position of the vanishing point in the image and the height of the indoor unit 1 held in advance, and further determines the direction of the corner. Then, the depth calculator 84 determines the depth of the room based on the calculated distance to the corner (89).

  After step 89, the space detection unit 29 outputs the position of the corner (distance to the corner, direction of the corner), the depth of the room, and the like (output 104). The detection unit 5 notifies the control unit 7 of the output 104.

  The control unit 7 controls the adjustment of air by the air conditioner based on the output 104. For example, the control unit 7 controls each device of the air conditioner so as to send a wind strengthened according to the depth of the room toward the corner calculated using the depth as a distance. Moreover, the control part 7 controls each apparatus of an air conditioner so that a wind direction swings according to the direction of a corner. Thereby, the air conditioner can appropriately adjust the air according to the floor plan of the room, and can adjust the room to a comfortable state.

  Below, the calculation method of the distance to the corner by the depth calculation part 84 is demonstrated.

  FIG. 18 is an explanatory diagram illustrating a calculation process of the distance to the corner by the depth calculation unit 84 according to the second embodiment.

  FIG. 18A is an explanatory diagram showing the vertical position of the vanishing point 99 in the image of the second embodiment.

  FIG. 18A is an image captured by the imaging unit 26. The image in FIG. 18A includes two vanishing points 99 (the vanishing point 99A and the vanishing point B). The depth calculation unit 84 acquires a vertical length a from the vanishing point 99 to the upper end of the image and a vertical length b from the vanishing point 99 to the lower end of the image based on the pixels of the image.

  FIG. 18B is an explanatory diagram illustrating the relationship between the vertical position of the vanishing point and the distance to the corner according to the second embodiment.

FIG. 18B shows a plane including the indoor unit 1 and a corner and parallel to the vertical direction. The imaging unit 26 illustrated in FIG. 18B captures an angle of view of 45 ° below the horizontal plane. For this reason, the relationship between the distance L on the horizontal plane from the indoor unit 1 to the corner, the length a and the length b in the image, and the height h at which the indoor unit 1 is installed is the ratio shown in the following Equation 4. Is represented by
L: h = (a + b): a (Formula 4)
Then, from Expression 4, the following Expression 5 is derived.
L = (a + b) × h / a (Formula 5)

  The depth calculation unit 84 can calculate the distance L using the length a and the length b, the height h at which the indoor unit 1 is installed, and Equation 5. When a plurality of vanishing points 99 are detected as shown in FIG. 18A, the depth calculation unit 84 calculates the distance L corresponding to each vanishing point 99. For example, the height h at which the indoor unit 1 is installed is 2 meters, and the depth calculation unit 84 holds the height h (2 meters) in advance.

  In the above-described example, the length a and the length b are acquired using an image when the imaging unit 26 images an upper and lower angle of view of 45 ° below the horizontal plane. However, as long as the depth calculation unit 84 holds an angle indicating the up and down direction that the imaging unit 26 captures, the orientation of the imaging unit 26 is not limited to 45 ° below the horizontal plane.

  For example, even when the imaging unit 26 captures an angle of view of 10 ° above and 35 ° below with respect to the horizontal plane, the depth calculation unit 84 is connected to the imaging unit 26 based on information on the upper and lower angles of the imaging unit 26. By specifying a position corresponding to the same height in the image, the length a can be obtained. In addition, the depth calculation unit 84 can calculate the distance L using Expression 5 if the vertical length (a + b) of the image captured by the imaging unit 26 is obtained instead of the length b.

  After calculating the distance L, the depth calculation unit 84 calculates the distance from the indoor unit 1 to the corner using the distance L on the horizontal plane, the height h, and the square theorem.

  Further, the depth calculation unit 84 may calculate the distance to the corner by using the result of the process shown in FIG. 17 performed on each of the plurality of images in the same direction. Specifically, the depth calculation unit 84 may calculate an average value of a plurality of results calculated in a plurality of processes shown in FIG. 17, and may determine the calculated average value as a distance to the corner. That is, the depth calculation unit 84 may learn the distance to the calculated corner, and may determine, for example, a distance obtained by averaging the latest 10 distance calculation results as the distance to the corner.

  Further, the depth calculation unit 84 is directly in front of the indoor unit 1 based on the horizontal position of the vanishing point 99 in the image and the direction (for example, left 32, front 33, or right 34 shown in FIG. 3). The direction of the corner is obtained by calculating the angle from to the corner. However, the depth calculation unit 84 may determine the direction of the corner using a method other than the method described above because the detection unit 5 can instruct the wind direction to the control unit 7 according to the determined direction.

  For example, the depth calculation unit 84 may obtain the direction of the corner by using the result of the process shown in FIG. 17 performed on each of the plurality of images in the same direction. Specifically, the depth calculation unit 84 may calculate an average value of a plurality of results obtained from the results of the processing shown in FIG. 17 a plurality of times, and may determine the calculated average value as a corner direction. . That is, the depth calculation unit 84 may learn the obtained corner direction, and may determine, for example, the direction obtained by averaging the latest 10 processing results as the corner direction.

  When the distance to the corner corresponding to the plurality of vanishing points 99 is calculated as shown in FIG. 18A, the depth calculation unit 84 determines the farthest distance as the depth of the room. For example, when the distance to the corner corresponding to the vanishing point 99B is longer than the distance to the corner corresponding to the vanishing point 99A, the depth calculation unit 84 determines the distance to the corner corresponding to the vanishing point 99B as the depth.

  Further, the depth calculation unit 84 may specify the coordinates of the vanishing point 99 in the image. And the detection part 5 may notify a coordinate to the control part 7 as information regarding a floor plan. Accordingly, when there are a plurality of corners included in the image, the detection unit 5 can notify the control unit 7 of information for controlling the wind direction so as to swing between the corners.

  Hereinafter, other methods for calculating the distance to the corner using the processing result of the human body detection unit 30 or the processing result of the temperature matrix detection unit 31 will be described.

  FIG. 19 is an explanatory diagram illustrating the processing result of the human body detection unit 30 according to the second embodiment and the distance and direction of the corner.

  FIG. 19 shows the relationship between the human body in the room detected by the human body detection unit 30 and the corner. FIG. 19 shows the position of the human body from the viewpoint of viewing the ceiling from the floor in the room. In step 89, the depth calculation unit 84 specifies the right corner direction 48 and the left corner direction 49 based on the image captured by the imaging unit 26. Then, based on the position of the human body detected by the human body detection unit 30, the depth calculation unit 84 estimates the distance from the indoor unit 1 to the corner.

  For example, the human body detection unit 30 regards the head of the human body as a circle, and calculates the distance from the indoor unit 1 to the human body based on the size of the circle corresponding to the head. Moreover, the human body detection part 30 calculates | requires the direction in which a human body exists, for example by calculating | requiring the direction of the circle corresponding to the head of a human body.

  For example, the depth calculation unit 84 regards the position of the rightmost human body 46 detected by the human body detection unit 30 as the position of the right wall, passes through the rightmost human body 46, and the indoor unit 1 is installed. A wall 101 that is perpendicular to the wall is virtually generated. Then, the depth calculation unit 84 determines the intersection of the generated wall 101 and the right corner direction 48 obtained from the image of the imaging unit 26 as the right corner. Then, the depth calculation unit 84 estimates the distance from the indoor unit 1 to the right corner and the position of the right corner by using the position and direction 48 of the human body 46, the height of the indoor unit 1, and the trigonometric function.

  The depth calculation unit 84 virtually generates a front wall 102 facing the indoor unit 1 in a direction parallel to the wall on which the indoor unit 1 is installed from the determined right corner, The intersection with the direction 49 of the left corner is determined at the left corner. Then, the depth calculation unit 84 uses the estimated position of the right corner, the direction 48 and the direction 49, the height of the indoor unit 1, and the trigonometric function to obtain a distance from the indoor unit 1 to the left corner. Is estimated.

  Further, the depth calculation unit 84 estimates the position of the farthest human body 47 in the direction of the front 33 as the position of the wall 103 facing the indoor unit 1, and sets the intersection of the estimated wall 103 and the direction 48 and the direction 49 to the right. Determine the corner and left corner. The depth calculation unit 84 estimates the distance from the indoor unit 1 to the right corner and the distance from the indoor unit 1 to the left corner by using the position of the human body 47, the direction 48, the direction 49, and the trigonometric function. To do.

  Then, the depth calculation unit 84 calculates a statistical value between the distance to the right corner estimated using the human body 46 and the distance to the right corner estimated using the human body 47, and the calculated statistical value To the distance to the right corner. In addition, the depth calculation unit 84 similarly calculates a statistical value for the estimated distance to the left corner, and determines the distance to the left corner. Here, the statistical value may be an average value, or a maximum value or a minimum value.

  Further, the depth calculation unit 84 estimates the distance to the right corner and the distance to the left corner using the human body located on the leftmost side, and uses all the estimation results to determine the distance to the right corner and The distance to the left corner may be determined. Further, the depth calculation unit 84 may learn the distance determined in this way, and may determine, for example, a value obtained by averaging the results of 10 times as the corner distance.

  Furthermore, as another method, for example, based on the processing result of the human body detection unit 30, the depth calculation unit 84 sets the farthest human body position in the direction of the front 33 as the position of the wall facing the indoor unit 1. The distance between the farthest human body and the indoor unit 1 may be calculated as the depth of the room. Further, the depth calculation unit 84 may learn the depth calculated a plurality of times, and may determine, for example, the average value of the latest 10 calculation results as the depth of the room.

  Furthermore, when the room in which the indoor unit 1 is installed is rectangular and the indoor unit 1 is installed in parallel to one wall side surface, the depth calculation unit 84 is based on the processing result of the human body detection unit 30 and The distance to the corner may be determined using the distance between the position of the human body located on the left side and the position of the human body located on the rightmost side and the distance from the indoor unit 1 to the human body located on the front. .

  Here, the distance between the position of the human body located on the leftmost side and the position of the human body located on the rightmost side is regarded as the distance between the corners at both ends of the wall facing the indoor unit 1. The distance from the indoor unit 1 to the human body located in front is regarded as the distance from the indoor unit 1 to the wall facing the indoor unit 1 and is regarded as the depth of the room.

  A method for calculating the depth, the distance to the corner, and the like based on the result detected by the temperature matrix detection unit 31 will be described below.

  The depth calculation unit 84 may specify the indoor floor based on the indoor surface temperature detected by the temperature matrix detection unit 31, and may calculate the distance to the corner using the specified floor.

  The floor temperature is affected by the temperature outside the room across the floor. For this reason, the depth calculation unit 84 specifies the position of the floor based on the difference between the room temperature detected by the temperature matrix detection unit 31 and the surface temperature.

  When the floor can be specified, the depth calculation unit 84 is installed with the above-described Equation 5, the ratio of the detected vertical field angle of the floor, and the indoor unit 1. Based on the height h, the depth to the wall facing the indoor unit 1 and the distance to the corner can be calculated. Further, the depth calculation unit 84 may learn the calculated depth, for example, calculate the distance from the depth and the direction of the corner to the corner by using the distance obtained by averaging the latest 10 detection results as the depth. .

  Further, based on the position of the human body detected by the human body detection unit 30, the depth calculation unit 84 may detect the position and direction of the corner of the space. Specifically, the depth calculation unit 84 assumes that the position of the leftmost human body among the human bodies detected by the human body detection unit 30 is the position of the left wall, and the position of the rightmost human body is the position of the right wall. Assuming the position, the position of the human body farthest in the front direction is assumed to be the position of the front wall, and the corner is specified by a combination of these assumed wall positions.

  Furthermore, the depth calculation unit 84 may learn the position or direction of the identified corner, and may determine, for example, the position or direction obtained by averaging the latest 10 specific results as the corner position or direction.

  Further, based on the indoor surface temperature detected by the temperature matrix detection unit 31, the depth calculation unit 84 may detect the position and direction of the corner of the space. Specifically, the position of the wall is specified based on the difference between the room temperature and the surface temperature, and further, the position or direction of the corner is specified based on the specified wall position.

  Further, the depth calculation unit 84 is configured to determine the distance and direction to the corner determined based on the image captured by the imaging unit 26, and the distance and direction to the corner determined based on the detection result of the human body detection unit 30. Etc., and the statistical values of the distance and direction to the corner obtained based on the detection result of the temperature matrix detection unit 31 are calculated, and the calculated statistical value is determined as the distance and direction to the corner. Good. The statistical value here may be, for example, an average value, a maximum value, or a minimum value.

  Furthermore, the space detection unit 29 may estimate the installation position of the indoor unit 1 using the image captured by the imaging unit 26, and the control unit 7 may adjust the wind direction based on the estimated installation position. For example, when the intersection in the image captured on the front surface 33 is biased to the right side, the space detection unit 29 may be estimated to be biased and installed on the right 34 side of the indoor unit 1. And according to an estimation result, the control part 7 may control the apparatus with which the indoor unit 1 and the outdoor unit 2 are provided so that a wind may be sent to 32 directions on the left.

  The indoor unit 1 described above was installed on the ceiling side of the room, and the imaging unit 26 was installed in the lower part of the indoor unit 1. However, the imaging unit 26 of the present embodiment may be installed in any direction as long as the room can be imaged widely. For example, when the indoor unit 1 is installed close to the floor, the lens of the imaging unit 26 May be installed in the upper part of the indoor unit 1.

  When the indoor unit 1 is installed in the lower part of the room, the depth calculation unit 84 may calculate the distance from the indoor unit 1 to the corner on the ceiling side. The depth calculation unit 84 calculates the distance to the corner on the ceiling side using Equation 5, and in this case, the vertical angle of view of the imaging unit 26 is 45 °, and the height h is from the indoor unit 1 to the ceiling. The length a is the length from the horizontal plane to the vanishing point on the ceiling side, and the length (a + b) is the vertical length of the image.

  According to the second embodiment, since the detection unit 5 can appropriately acquire the room layout (corner position (distance, direction), depth, etc.), the control unit 7 according to the second embodiment has a shape and size of the room ( The air volume or direction can be adjusted appropriately according to the floor plan. As a result, the air conditioner of Example 2 can be adjusted to a comfortable state in the room.

  Further, by detecting the floor plan using the image from which the distortion is removed by the distortion removing unit 290 according to the second embodiment, for example, even when the imaging unit 26 having a wide-angle lens captures a wide-angle image, an image with a narrow viewing angle is obtained. Thus, it becomes possible to accurately detect beams and the like that were not reflected.

  In addition, this invention is not limited to an above-described Example, Various modifications are included.

  For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described. Further, a part of the configuration of one embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of one embodiment. In addition, it is possible to add, delete, or replace another configuration for a part of the configuration of each embodiment.

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

  Further, the control lines and information lines indicate objects that are considered necessary for the explanation, and not all control lines and information lines on the product are necessarily shown. Actually, it may be considered that almost all the components are connected to each other.

DESCRIPTION OF SYMBOLS 1 Indoor unit 2 Outdoor unit 3 Remote control 4 Sensor 5 Detection part 7 Control part 11 Electrical component 26 Imaging part 27 Temperature detection part 29 Space detection part 30 Human body detection part 31 Temperature matrix detection part

Claims (6)

An air conditioner,
An imaging unit that captures an indoor image;
A storage unit for storing the captured image;
A region for removing distortion from the captured image, and a conversion unit for converting the specified region;
A detection unit that detects features in the room based on an image including the converted area;
An air conditioner, comprising: a control unit that adjusts an amount or direction of air sent to the room based on the indoor characteristic detected by the detection unit. The air conditioner according to claim 1,
The air conditioner characterized in that the conversion unit converts the specified area by generating a planarized area from which distortion is removed from the specified area. The air conditioner according to claim 2,
The conversion unit performs image conversion using a normal image conversion unit and an interpolation processing unit,
The normal image conversion unit converts the corresponding point in the specified area into a corresponding point in the planarized area by performing a normal image conversion,
The air conditioner, wherein the interpolation processing unit generates pixels in the planarization region by performing cubic interpolation on pixels around corresponding points in the planarization region. The air conditioner according to claim 1,
The imaging unit captures a plurality of images in the room,
The air conditioner further includes a difference specifying unit that specifies a difference area between the plurality of images,
The said conversion part specifies the area | region of the specified difference as an area | region which removes the said distortion, The air conditioner characterized by the above-mentioned. The air conditioner according to claim 1,
The imaging unit acquires a first image by imaging the room, acquires the first image, and then acquires a second image by imaging the room;
The detection unit detects an area including a room partition by specifying at least one difference area between the first image and the second image, and detects the detected area in the second image. By calculating the complexity of the area, detecting whether the partition included in the detected area is in an open state or a closed state in the second image as an indoor feature. A featured air conditioner. The air conditioner according to claim 1,
The air conditioner characterized in that the image pickup unit picks up the image with a wide-angle lens having a horizontal viewing angle of 150 degrees.