KR20120036139A - Air conditioner and method for sensing human - Google Patents

Abstract

PURPOSE: An air conditioner and a human body detecting method thereof are provided to detect persons inside a room and exactly circulate a position of the persons so that the accuracy of the position and detection of the persons inside the room are improved. CONSTITUTION: An air conditioner comprises an image photographing unit, a human body detecting unit, a position deciding unit, and a controlling unit. The image photographing unit takes a photograph of an indoor area. The human body detecting unit detects a movement inside the indoor area based on data generated through analyzing recorded image data per predetermined time interval by the image photographing unit and sets a human body detecting area, and decides the existence of persons inside the room with respect to the human body detecting area so that the persons inside the room is detected. The position deciding unit detects the position of the persons inside the room in the indoor area based on coordinate information with respect to the persons inside the room detected by the human body detecting unit. The control unit controls an air current discharged by corresponding to the position information of the persons in the room of the position deciding unit and a result of detecting the persons inside the room by the human body detecting unit.

Description

Air conditioner and method for sensing human

The present invention relates to an air conditioner and a human body sensing method. In particular, the present invention relates to an air conditioner and a human body sensing method for controlling an operation by detecting an occupant and determining its position through an image photographing means.

The air conditioner is installed to provide a more comfortable indoor environment for humans by discharging cold air into the room to adjust the indoor temperature and purifying the indoor air to create a comfortable indoor environment. In general, an air conditioner includes an indoor unit which is configured as a heat exchanger and installed indoors, and an outdoor unit which is configured as a compressor and a heat exchanger and supplies refrigerant to the indoor unit.

The air conditioner is separated and controlled by an indoor unit composed of a heat exchanger and an outdoor unit composed of a compressor and a heat exchanger, and is operated by controlling power supplied to a compressor or a heat exchanger. In addition, the air conditioner may be connected to at least one indoor unit to the outdoor unit, the refrigerant is supplied to the indoor unit according to the requested operating state, the operation is operated in the cooling or heating mode.

Recently, the air conditioner is equipped with means for capturing an image, and a method of checking whether the room is occupied has been proposed. However, the user can feel comfort because only simple control of the indoor unit can be performed by simply determining the indoor room. There is a limit to this.

In addition, in the case of three-dimensional occupant determination using a plurality of image capturing means, a plurality of image capturing means must be provided, and thus a manufacturing cost increases, and a processor for processing a 3D image is required. There is a problem that fast detection is difficult.

An object of the present invention is to enable a more accurate position determination in detecting the occupants through the image recording means, and to control the operation in consideration of the number of occupants detected, the air conditioner to make the user feel comfortable and its The present invention provides a method for detecting a human body.

An air conditioner according to the present invention includes an image photographing unit for photographing an indoor area; Based on the difference image data generated by analyzing the image data photographed at predetermined time intervals from the image capturing unit, the motion within the indoor area is detected and set as the human body area, and whether the human body is included in the human body area is determined. Human body detecting unit for detecting; A position determining unit detecting the position of the occupant in the indoor area based on the coordinate information of the occupant detected by the human body detecting unit; And a controller configured to control the airflow discharged in response to the occupant detection result of the human body detecting unit and the position information of the occupant of the position determining unit.

In addition, the human body detection method of the air conditioner of the present invention comprises the steps of generating the difference image data by comparing the image data taken the indoor area at a predetermined time interval; Setting a human body area by detecting a motion from the difference image data; Applying a skin filter to the human body region to finally determine whether a human body is present and detecting an occupant; Detecting position information of the human body region based on coordinate data when the human body region is determined to be a human body; And controlling the airflow based on the positional information.

The air conditioner and the human body detecting method according to the present invention configured as described above detect the occupant through the image capturing means and calculate the position accurately, thereby improving the accuracy of detecting the occupant and the position.

In addition, by operating in consideration of the number of detected occupants, a plurality of occupants can feel comfortable to provide a more comfortable indoor environment. Accordingly, the user's satisfaction is greatly improved.

1 is a perspective view showing a main body configuration of an air conditioner according to an embodiment of the present invention.
2 is a block diagram showing a control configuration of an air conditioner according to an embodiment of the present invention.
3 is a view referred to for explaining an image processing method for detecting a human body in an air conditioner according to an embodiment of the present invention.
4 is a view referred to for explaining an image processing method for detecting a plurality of human body when detecting the human body of the air conditioner.
FIG. 5 is a diagram referred to for describing a method of calculating a detected position of a human body.
FIG. 6 is a diagram referred to for describing a specific method according to calculating a human body position of FIG. 5.
7 is a flowchart referred to for explaining a human body sensing method of the air conditioner according to an embodiment of the present invention.
FIG. 8 is a flowchart of a method of processing a photographed image for detecting a human body among the human body detecting method of FIG. 7.
9 is a flowchart referred to for explaining a method of estimating a human body region.
FIG. 10 is a flowchart referred to for describing a method of finally determining whether a human body is present in a section estimated as the human body region.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

1 is a perspective view showing a main body configuration of an air conditioner according to an embodiment of the present invention.

As shown in Figure 1, the air conditioner of the present invention, the air inlet is formed in the lower portion, the air discharge port (8, 14) is formed in the upper, after air conditioning the air sucked through the air inlet in the air It has a flow path which discharges air through the discharge ports 8 and 14.

In addition, the main body 2 includes an indoor fan (not shown) for generating a blowing force to suck the indoor air into the main body 2 and discharge it to the outside of the main body 2, and the air blown from the indoor fan. It includes an indoor heat exchanger (not shown) for heat exchange with the refrigerant.

In this case, the air conditioner main body 2 may be applied to any case such as a stand-type air conditioner, a wall-mounted air conditioner, or a ceiling-type air conditioner, but will be described below by taking a stand-type air conditioner as an example.

The main body 2 is provided with vanes 96 and 81 for guiding air while opening and closing at least one of the air intake port and the air discharge port 8 and 14. Hereinafter, the vanes 81 and 96 simultaneously open and close the air inlet and the air outlet 8 and 14 together to guide both the intake air and the discharge air, and are formed to be long in the vertical direction as a whole.

At this time, the upper portion of the main body 2 is provided with an air discharge port on the left and right, respectively. The air discharge ports on the left and right sides are formed to be opened to the left and right, or are formed to be opened in the left or right front direction.

In addition, the main body 2 is formed with a discharge port unit having an air discharge port 14 formed therein, the discharge port unit being lifted above the main body 2 from the inner upper part of the main body 2 and above the main body 2 of the main body 2. It has a discharge part structure descending inward.

At this time, the upper portion of the main body 2 is provided with an image capturing unit 130 as an image capturing means for capturing an indoor image. The image photographing unit may be provided at any one of the top, bottom, left, right and center portions of the front panel of the main body 2. In addition, as shown, the image capturing unit 130 may be provided in the discharge port unit to be elevated upward from the main body (2).

In addition, the main body 2 may include an input unit for inputting a user command and an output unit for displaying an operation state.

2 is a block diagram showing a control configuration of an air conditioner according to an embodiment of the present invention. As shown in FIG. 2, the air conditioner includes an image capturing unit 120 provided as described above, a human body detecting unit 130, a position determining unit 140, a data unit 150, an input unit 160, The output unit 170, the wind direction control unit 180, the fan control unit 190 and the controller 110 for controlling the overall operation of the main body (2).

The data unit 150 stores control data used for driving the air conditioner, screen configuration data output through the output unit 170, and sound effect data. Also, the image signal captured by the image capturing unit 120 is stored. Temporarily stored or converted image data is stored. In addition, the reference data according to the human body monitoring through the image data is stored, the coordinate data for reading the location of the occupant is stored.

The input unit 160 is provided with at least one data input means through a pressing operation or a touch, and applies the input data to the controller 150. The input unit 160 receives a user command and setting data according to the operation of the air conditioner, receives a driving command, or an operation stop command and applies it to the controller 110.

The output unit 170 includes display means capable of outputting letters, numbers, and symbol images, and displays an operation state of the air conditioner on the screen. In this case, the output unit 170 may further include a speaker or a lamp, and outputs a warning sound or an effect sound according to the operation state, or outputs an operation state by turning on or flashing the lamp.

The fan controller 190 controls the amount of air discharged to the room by controlling a motor connected to the indoor fan in response to a control command of the controller 110.

The wind direction controller 180 adjusts the direction of the air discharged into the room in response to the control command of the controller 110, so that the discharged air is changed up, down, left, and right when the discharge port is opened. At this time, the wind direction control unit 180 includes a left wind direction control unit, a right wind direction control unit, and also includes an upper wind direction control unit of the upper discharge port unit.

The image capturing unit 120 includes a lens and an image sensor for converting light incident through the lens into an electrical image signal. In this case, the image sensor may be a metal oxide semiconductor (CMOS), a charge coupled device (CCD), but is not limited thereto.

As described above, the image capturing unit 120 is provided on any one of up, down, left, and right sides of the main body 2 to photograph an indoor image. In addition, it may be attached to the main body 2 as a separate device, or may be installed around the main body 2, and in this case, the image capturing unit 120 is connected to the main body 2 by wire or wirelessly.

In addition, the image capturing unit 120 is provided with a rotation means, it is possible to rotate the left and right within a predetermined angle range to take an indoor image. Here, the image capturing unit 120 photographs a two-dimensional indoor image, and the photographed image signal is input to the human body detecting unit 130.

The human body detecting unit 130 generates image data by processing the image signal input from the image capturing unit 120 and converting the image signal into a readable digital signal, and detects the human body from the image data. In this case, the human body detector 130 may calculate the amount of light and the frame rate of the image signal from the image signal, and extract color information for each pixel of the image data.

The human body detecting unit 130 detects the human body using the image data stored in the data unit 150 or a plurality of image data photographed at predetermined time intervals. In other words, the human body detection unit 130, the position determination unit 140 may detect the presence of the occupant by extracting the image area that is changed from the accumulated image data.

The human body detecting unit 130 compares the photographed image data with reference data stored in the data unit 150 or image data photographed before a predetermined time, detects an inconsistent image region, and uses the same to determine the presence of an occupant. . In addition, the human body detecting unit 130 may extract the changing image area from the accumulated image data to detect the presence of the occupants.

In addition, the human body detecting unit 130 detects the number of the human body, that is, the number of occupants from the image data.

The position determining unit 140 detects the position of the human body when the human body is detected from the image data by the human body detecting unit 130, that is, when a patient is present in the room. The position determination unit 140 detects the position of the human body, that is, the occupant, with reference to the coordinate data stored in the data unit 150.

The position determination unit 140 may detect the position using the stored coordinate data. The position determination unit 140 may use the generated coordinate data or generate the coordinate data separately to use the position detection.

At this time, the coordinate data is the coordinate data of the indoor area where the air conditioner main body 2 is located, and is a reference data for estimating the actual position of the indoor area based on the coordinates in the image.

The control unit 110 receives the presence detection result of the occupants, the number of occupants and the position information of the occupants from the human body detecting unit 130 and the position determining unit 140, and sets the operation of the indoor setting temperature, wind direction, wind speed, and the like. Change and control.

The controller 110 variably controls the temperature of the air discharged into the room according to the changed setting, and controls the fan controller 190 to variably control the wind direction or the wind speed. In addition, the wind direction control unit 180 is controlled to change the wind direction in accordance with the location of the occupant.

Here, the controller 150 may change the operation setting for the set temperature and the wind speed with reference to the comfort index proposed by Ashrae. For example, the set temperature is preferably 23 to 26, and the wind speed is maintained at 0.15 m / s or less.

3 is a view referred to for explaining an image processing method for detecting a human body in an air conditioner according to an embodiment of the present invention.

Referring to FIG. 3, when the image capturing unit 120 captures an indoor image according to a control command of the controller 110, the human body detecting unit 130 captures an image input from the image capturing unit 120 according to a predetermined codec. Convert and generate image data.

The human body detector 130 compares the first image data shown in FIG. 3A with the second image data shown in FIG. 3B. The human body detecting unit 130 generates difference image data as shown in FIG. 3C for the difference between the first image data and the second image data.

In this case, the first image data is image data that is a target for detecting a human body as the captured image data.

Image data photographed at the second time. The second image data is image data photographed before the reference data or the first image data. That is, when the second image data is photographed at the first time t-1, the first image data is data photographed at the second time t later than the first time. In this case, the second image data may be image data that previously detected the human body.

The human body detecting unit 130 generates difference image data by using an absolute value obtained by subtracting the second image data from the first image data. In this case, the fixed object is photographed in the same manner, and thus the difference value is left as an image for the moving object or the human body.

This is expressed as a formula as follows.

In this case, when the difference is greater than or equal to a predetermined reference value, the difference image data is generated. In other words, if there is a difference but the area is very small compared to the entire frame, it is regarded as the same image data and judged that there is no motion.

Accordingly, the human body detecting unit 130 detects a moving object or a human body from the image data by using the difference image data comparing the two image data.

The human body detecting unit 130 extracts color information from the generated image data and finally determines whether the detected object is an animal, an object, or a human body to detect the human body.

The human body detection unit 130 distinguishes the human body by applying a skin filter to the difference image data, that is, determines whether the human body is the color of the difference image data region by the color of the skin.

At this time, the human body detection unit 130 applies a skin filter having the following conditions, and if the condition is satisfied, it is determined as the skin color. However, the skin color varies according to race, so the figures may change accordingly. The following describes the yellow race as an example of determining the skin color. In addition, the color information will be described as an example of the RGB method.

The human body detecting unit 130 detects a moving object through the difference image data and applies the skin filter to the image region of the difference image data to determine whether the object to be moved is an object, an animal, or a human body. Detect it.

4 is a view referred to for explaining an image processing method for detecting a plurality of human body when detecting the human body of the air conditioner.

Referring to FIG. 4, the human body detecting unit 130 detects the number of human bodies, that is, the number of occupants from the captured image data.

The human body detector 130 accumulates the difference image data and determines the number of human bodies based on the accumulated value for a predetermined time.

FIG. 4A is summation data of difference image data, FIG. 4B is a distribution of lateral cumulative sums, and FIG. 4C is a distribution of longitudinal cumulative sums.

The human body detector 130 generates difference image data, respectively, as shown in FIG. 3 with respect to the plurality of image data photographed for a predetermined time, and accumulates and sums them as shown in FIG.

At this time, the cumulative sum in the horizontal direction is calculated with respect to the sum data of the difference image data accumulated and summed as shown in b of FIG. 4, and the cumulative sum is calculated in the longitudinal direction as shown in c of FIG. 4.

That is, in the sum data of the difference image data, the cumulative sum of the portions where the actual pixels appear in the lateral direction and the cumulative sums of the portions where the actual pixels appear in the longitudinal direction are shown.

As a result, an area where a large number of pixels are distributed in the longitudinal direction and the horizontal direction can be detected, and the number of the human body is detected based on the vertical and horizontal pixel distribution areas.

In the case of Fig. 4, two regions appear in the transverse direction and four regions in the longitudinal direction. In this case, one of the two regions detected in FIG. 4b has a small peak value and the other one appears large, and thus, more people are located in the second region than in the first region. When the peak values are compared, it can be seen that the peak value of the second region is three times the peak value of the first region, that is, the difference in the number of people is three times.

In addition, referring to c of FIG. 4, the human body appears in four regions in the cumulative sum distribution in the longitudinal direction.

Accordingly, the human body detecting unit 130 may determine that four human bodies are divided into two groups and exist in the room.

FIG. 5 is a diagram referred to for describing a method of calculating a detected position of a human body. 5 is a diagram illustrating an example of coordinate data as a reference for calculating a human body position.

The position determining unit 140 calculates the coordinates of the region determined as the human body based on the result detected by the human body detecting unit 130 using the coordinate data as shown in FIG. 5.

In this case, the coordinate data is data for providing a criterion for calculating the position of the human body with respect to an area where the human body is to be detected, that is, an indoor area where the main body is installed.

Such coordinate data may be attached to a plurality of markers representing coordinates in an indoor area as shown, and may be used as coordinate data by photographing an indoor area to which a marker is attached.

In this case, in capturing the coordinate data, the plurality of markers are attached from the main body in the shape of a fan at intervals to display the distances, and the markers M are attached to the left and right at regular angular intervals to represent angles relative to the main body.

Even if a plurality of markers are uniformly requested according to a predetermined distance and a predetermined angle, the coordinate data is two-dimensional planar image data of photographing them, so that the intervals between the markers of the plurality of photographed markers are not constant.

That is, the interval b according to the distance from the main body means that the intervals are different in FIG. 5, but the same distance in the actual indoor area. In addition, the left and right intervals of the angle markers represent the left and right angles, and thus the angles may be different from each other in FIG.

Since the coordinate data includes coordinate information about each photographed marker, the position determination unit 140 calculates a position through coordinate information of the marker according to which marker the region of the human body is located.

FIG. 6 is a diagram referred to for describing a specific method according to calculating a human body position of FIG. 5.

When the occupant is located in one area of the room, the whole body may be photographed from head to toe according to the position of the occupant, and in some cases, only a part of the body may be photographed.

In this case, when the image capturing unit 120 photographs the indoor area as shown in FIG. 6, the human body detecting unit 130 converts the image input from the image capturing unit 120 to the image capturing area. Create and analyze it to detect the human body.

Accordingly, the human body is detected in the area where the occupant is photographed from the image data, and the position determining unit 140 detects the position of the area determined as the human body.

The position determining unit 140 determines the end of the image region determined as the human body from the photographed two-dimensional image data, that is, the lowermost end and the central portion of the image region with respect to the human body as the position of the human body. That is, it is determined that the human body is located at the first point P1 of FIG. 6.

Accordingly, the position determination unit 140 detects the coordinate information of the marker with respect to the first point P1 and the human body region around the position based on the coordinate data, and calculates the position of the occupant therefrom.

For example, when the first point P1 is the sixth marker in distance from the main body and corresponds to the fifth marker from the left, when the markers are attached at intervals of 30 cm and 10 degrees, 50 degrees from the left side of the main body and 180 cm from the main body It can be determined that the location is located at.

At this time, when the human body region is difficult to specify the position of the first point P1 with a single marker, that is, when the human region is located in the middle of the plurality of markers, the position is determined using the ratio of the distance between the markers and the actual marker. Can be calculated.

The controller 110 detects the human body by the human body detector 130, and when the position is detected by the position determination unit 140, the controller 110 controls the airflow to the point where the occupant is located. Depending on the setting, direct wind or indirect wind is provided.

7 is a flowchart referred to for explaining a human body sensing method of the air conditioner according to an embodiment of the present invention.

Referring to FIG. 7, the controller 110 drives the image capturing unit 120 according to a set mode to start detecting a human body. The image capturing unit 120 captures an image of an indoor area according to a control command of the controller 110 (S210).

The human body detector 130 detects a movement from the image data by converting the photographed image (S220).

In addition, the human body detecting unit 130 sets a section for the region where the motion is detected, and estimates the set section as the human region (S230).

The human body detecting unit 130 applies a skin filter to a region estimated to be a human body region, and determines the human body for each region according to the result (S240). In this case, as a result of applying the ski filter in the corresponding area, the human body detecting unit 130 determines that the human body is the skin color and the animal or the object when the skin color is not the skin color.

When the human body is detected by the human body detecting unit 130, the position determining unit 140 detects the coordinates of the human body region based on the preset coordinate data (S250). The position determining unit 140 calculates an actual position in the indoor area with respect to the detected coordinates (S260). At this time, the position determination unit 140 uses the angle and the distance value, and the coordinate data includes information about the position in the actual area.

When the human body is sensed by the human body detecting unit 130, the controller 110 determines that the occupant is present in the indoor area, and controls the airflow based on the position information of the occupant detected by the position determining unit 140. (S270). In this case, the controller 110 controls indirect wind or direct wind according to the set mode.

FIG. 8 is a flowchart of a method of processing a photographed image for detecting a human body among the human body detecting method of FIG. 7. The human body detecting method of the human body is described as follows.

Referring to FIG. 8, the image capturing unit 120 captures an image of an indoor area (S310). In this case, the image capturing unit 120 may repeatedly photograph the indoor area at predetermined time intervals according to a control command of the controller 110.

The human body detector 130 compares the first image data with the second image data and stores color information for each pixel of each image data (S320). In this case, the human body sensing unit 130 will be described with an example of storing color information of each image data in an RGB method. The RGB method is a method of displaying a color represented by a combination of three primary colors (red, green, and blue) by a certain number of three primary colors.

The human body detecting unit 130 calculates the absolute difference with respect to the color information per pixel (S330). The human body detecting unit 130 calculates the absolute difference between the matched pixels by matching the color information per pixel of the first image data and the color information per pixel of the first image data, in units of pixels.

The human body detector 130 detects a change in the image over time by comparing the image data photographed at different times, that is, the first image data and the second image data.

In this case, when using the RGB color information, the human body detection unit 130 has three color values for each pixel as color information, so that the first image data and the second image data are compared in pixel units, respectively. Calculate the absolute differences for each of the three color values.

When the absolute difference of the color information is calculated, the human body detecting unit 130 determines whether there is a pixel whose absolute difference exceeds a reference value according to the calculation result (S340).

Since the human body detector 130 has three color values per pixel, the human body detector 130 calculates three absolute differences, and determines whether at least one of the three absolute differences per pixel exceeds the reference value.

If there is a pixel whose absolute difference exceeds the reference value, the human body detecting unit 130 changes the pixel to white (S350), and if the absolute difference is less than or equal to the reference value, the pixel changes to black (S360).

The human body detecting unit 130 calculates the difference between the color information of the pixels of the first image data and the second image data by comparing the reference values of the absolute difference as described above, and according to the result, the difference displayed in white and black. The image data is generated (S370).

In this case, when the matched pixels of the first image data and the second image data are the same, the difference becomes 0, and thus, the corresponding pixel becomes black in the difference image data, and when the color information of the matched pixel is not the same, the range is similar. That is, when the reference value is less than the reference value, it is displayed in black, and when the color information of the pixel is different than the reference value, it is displayed in white. Accordingly, difference image data as shown in FIG. 3C is generated.

FIG. 9 is a flowchart referred to for describing a method of estimating a human body region from difference image data of FIG. 8.

Referring to FIG. 9, the human body detecting unit 130 designates a region of interest for a column at a first size from the difference image data generated as described above (S410). In addition, the human body detection unit 130 designates a region of interest for each row in the second size from the difference image data. In this case, as illustrated in FIG. 4, the ROI may be uniformly set in the shape of a rectangle.

For example, a region of interest may be set by specifying a region of interest in 1 × horizontal size, and a region of interest in columns may be designated by specifying a region of interest in 1 × vertical size. In this case, the region of interest of 320 columns and the region of interest of 240 rows may be generated for image data having a 240x320 resolution.

In some cases, the region of interest may be partially specified only for a portion of the difference image data based on the region marked with white, and in the image processing, the region of interest may be extracted by extracting a shape similar to an edge shape or a human body shape. It may be.

If the region of interest is specified, the human body detecting unit calculates the pixel sum for each region of interest separately (S430). At this time, since the difference image data is a binary image of white and black, each pixel value is 0 or 255.

The human body detector 130 generates an array of pixel sums (S440), and calculates an average value in a predetermined unit in the array (S450).

For example, as described above, in the image data having 240x320 resolution, an array including a pixel sum for 320 columns and a pixel sum for 240 rows is generated. In addition, the average value may be sequentially calculated in units of 5 units. In this case, the predetermined unit may be changed according to the resolution of the image data and the accuracy of the human detection to be determined.

The average value thus calculated is changed to the representative value for the numerical values of the arrays (S460). That is, if the average value is calculated in five units in the array, the average value calculated in place of the five numbers is replaced.

If it is changed to an array of average values, it is determined whether the representative value (average value) constituting the array is larger than the set value (S470).

At this time, if the representative value (average value) is larger than the set value, the section is stored according to the row and column, and the corresponding section is estimated as the human body region (S480). On the other hand, if the representative value is smaller than the set value, it is determined that the non-human area.

As described above, the motion of the image data may be detected using the average value of the pixel sum, and the human body region may be estimated based on the motion.

In addition, by setting an area in which the human body is estimated to exist, when a plurality of occupants are present in the indoor area when the human body is detected, the number of the plurality of occupants may be detected.

FIG. 10 is a flowchart referred to for describing a method of finally determining whether a human body is present in a section estimated as the human body region.

As described above, when a section estimated to be the human body region is set, it is finally determined whether the region is the region where the human body is actually located.

The human body detecting unit 130 stores color information for each pixel in the image data for the section set as the human body region (S510).

The human body detecting unit 130 applies a skin filter to the set section (S520), and compares the skin condition according to the skin filter setting in units of pixels (S540).

In this case, the skin filter is for determining whether the skin color is a pixel unit for the corresponding section. The skin color can be changed and set according to the race of the occupants to be photographed, and a plurality of colors can be designated as the skin color.

The skin filter may be set under the condition of Equation 1 and 2 described above.

The human body detector 130 determines whether the skin condition of the skin filter is satisfied on a pixel-by-pixel basis (S540), and when the pixel value satisfies the skin condition, the body value 130 is set to white (S550). On the other hand, if the pixel value does not satisfy the skin condition, the pixel value is designated as black (S560).

At this time, white has a value of 255 black.

That is, the human body detecting unit determines whether a skin color exists in a corresponding section by applying a skin filter to the estimated region of the human body. In this case, when the skin color is not white, the body color is changed to black.

When the skin filter is changed to white and black data by applying a skin filter to the pixel as described above, the human body detector 130 calculates a sum of each pixel value for each section which is a human body region (S570).

That is, when there are a plurality of sections estimated as the human body area, a sum of pixel values is calculated for each section.

The human body detecting unit 130 determines whether the sum of the pixel values is 0 (S580), and if the pixel sum is black, determines that the human body is not (S600). (S590).

If a skin color exists even in one section, the corresponding pixel value is 255. Therefore, since the pixel sum of the section is not zero, it is determined that a human body exists in the section.

On the other hand, if the pixel sum is 0, it means that no pixel satisfies the condition of the skin filter in the section, and thus it is determined that the human body does not exist.

The human body detector 130 calculates difference image data from the image data as described above, estimates the human body region therefrom, and finally determines whether the human body is using a skin filter.

The position determination unit 140 that detects the human body calculates the position of the human body using coordinate data as described above with reference to FIGS. 5 and 6, and the controller 110 controls the airflow based on the position of the occupant.

Accordingly, the present invention can not only detect the human body by a simple movement, but can accurately distinguish between an object or an animal and a human body, and thus more accurately detect a patient. In addition, it is possible to provide a more comfortable environment through accurate human detection and position determination as described above.

As described above, the air conditioner and the human body detecting method according to the present invention have been described with reference to the illustrated drawings. However, the present invention is not limited by the embodiments and the drawings disclosed herein, and the technical idea is protected within the scope. Can be applied.

2: body
110: control unit 120: video recording unit
130: human body detection unit 140: position determination unit
150: data section
180: wind direction control unit 190: fan control unit

Claims (17)

An image photographing unit photographing an indoor area;
Based on the difference image data generated by analyzing the image data photographed at predetermined time intervals from the image capturing unit, the motion within the indoor area is detected and set as the human body area, and whether the human body is included in the human body area is determined. Human body detecting unit for detecting;
A position determining unit detecting the position of the occupant in the indoor area based on the coordinate information of the occupant detected by the human body detecting unit; And
And a controller configured to control the airflow discharged in response to the occupant detection result of the human body detecting unit and the position information of the occupant of the position determining unit. The method of claim 1,
And the human body detecting unit matches the pixels of the first image data photographed at a first time and the second image data photographed at a second time one-to-one, and compares color information of the pixels to generate the difference image data. The method of claim 2,
The human body detecting unit calculates an absolute difference with respect to the color information of the pixel, compares the absolute difference with a preset reference value, and changes the corresponding pixel to white or black according to whether the reference value is exceeded. Air conditioner for generating car image data. The method of claim 1,
The human body detecting unit sets a plurality of ROIs having a predetermined size in the difference image data, detects motion from the difference image data by calculating pixel sums and average values of the plurality of ROIs, and detects a region where the movement is detected. An air conditioner for setting the human body area. The method of claim 4, wherein
The human body detector generates an array of pixel sums by calculating a pixel sum for each region of interest, reconstructs the array by using an average value calculated in a predetermined unit with respect to the array as a representative value, If the sum is greater than or equal to the set value, the air conditioner determines that there is movement and stores the row and column sections for the region of interest and sets it as the human body region. The method of claim 1,
The human body detecting unit extracts color information of the human body region from the image data, and applies a skin filter to the human body region to finally determine whether the human body is an air conditioner. The method according to claim 6,
The human body detecting unit detects a pixel satisfying a condition for a skin color designated by at least one color through the skin filter, and when at least one pixel satisfying the condition is detected in the human body, the object of the human body is a human body. Air conditioner judged to be. The method of claim 1,
The human body detecting unit detects at least one human body region from the difference image data, and detects the presence of occupants and the number of occupants.
The address setting button is an air conditioner comprising at least one input means of a button, a switch, and a touch pad. The method of claim 1,
The position determining unit detects the position information of the occupant by extracting coordinate information based on the center of the lower end of the human body area when compared to the preset coordinate data when the occupant is detected by the human body detecting unit. The method of claim 9,
The position determining unit based on the coordinate data stored by photographing the indoor area attached with a plurality of markers at predetermined intervals, the position based on the coordinate information on the marker near the point where the center of the lower end of the human body area is located Air conditioner to detect information. The method of claim 1,
The control unit controls an airflow by changing at least one of an operation mode, a wind direction, a wind speed, and a temperature based on the presence of occupants detected by the human body detecting unit, the number of occupants, and the position information of the position determining unit. group. Generating difference image data by comparing image data photographed at an indoor area at predetermined time intervals;
Setting a human body area by detecting a motion from the difference image data;
Applying a skin filter to the human body region to finally determine whether a human body is present and detecting an occupant;
Detecting position information of the human body region based on coordinate data when the human body region is determined to be a human body; And
Human body sensing method of the air conditioner comprising the step of controlling the air flow based on the position information. The method of claim 12,
Calculating an absolute difference with respect to color information for each pixel of the image data, and comparing the absolute difference with a preset reference value to generate the difference image data composed of white and black. The method of claim 12,
When the absolute difference is calculated for each of the three color values included in the color information, the pixel is changed to white when at least one absolute difference exceeds a preset reference value, and when all absolute differences do not exceed the reference value. Human body detection method of air conditioner to change to black. The method of claim 13,
A plurality of ROIs are set in the difference image data to calculate an pixel sum for each ROI, and an array is calculated. The average value is calculated in a predetermined unit to reconstruct the array into an array having the average value as a representative value. If the sum of the elements of the array is equal to or more than the set value is determined that there is a movement, the human body sensing method of the air conditioner to store the row and column section for the region of interest set to the human body region. The method of claim 13,
The coordinate data is data of photographing the indoor area in which a plurality of markers are attached at predetermined intervals, wherein the human body sensing method of the air conditioner includes coordinate information about the plurality of markers. The method of claim 13,
The human body detecting method of the air conditioner for extracting the position information of the occupant by extracting the coordinate information of the location of the center of the lower end of the human area.

Images