CN114120354A - Human body detection and positioning method and device applied to air conditioner and intelligent sensing system - Google Patents
Human body detection and positioning method and device applied to air conditioner and intelligent sensing system Download PDFInfo
- Publication number
- CN114120354A CN114120354A CN202111199532.9A CN202111199532A CN114120354A CN 114120354 A CN114120354 A CN 114120354A CN 202111199532 A CN202111199532 A CN 202111199532A CN 114120354 A CN114120354 A CN 114120354A
- Authority
- CN
- China
- Prior art keywords
- human body
- image
- air conditioner
- target
- distance
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 41
- 238000000034 method Methods 0.000 title claims abstract description 37
- 230000011218 segmentation Effects 0.000 claims abstract description 38
- 238000012216 screening Methods 0.000 claims abstract description 24
- 238000013507 mapping Methods 0.000 claims abstract description 17
- 238000004378 air conditioning Methods 0.000 claims abstract description 13
- 238000010191 image analysis Methods 0.000 claims description 16
- 238000009529 body temperature measurement Methods 0.000 claims description 15
- 238000012937 correction Methods 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 8
- 230000005540 biological transmission Effects 0.000 claims description 6
- 230000036760 body temperature Effects 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 4
- 230000008447 perception Effects 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 230000007797 corrosion Effects 0.000 claims description 3
- 238000005260 corrosion Methods 0.000 claims description 3
- 230000003628 erosive effect Effects 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims description 3
- 230000006870 function Effects 0.000 description 15
- 238000012545 processing Methods 0.000 description 13
- 230000000694 effects Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 230000036544 posture Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
- F24F11/63—Electronic processing
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F2120/00—Control inputs relating to users or occupants
- F24F2120/10—Occupancy
- F24F2120/12—Position of occupants
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Air Conditioning Control Device (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a human body detection and positioning method, a human body detection and positioning device and an intelligent sensing system applied to an air conditioner, wherein the method comprises the following steps: carrying out threshold segmentation on a human body infrared image acquired in an air-conditioning scene; extracting and tracking target edges of the image subjected to threshold segmentation to obtain a plurality of complete contours; carrying out circumscribed rectangle description on each complete outline to form a plurality of rectangular target frames; screening a plurality of rectangular target frames through human body constraint; determining the orientation of the human body target to be detected relative to the air conditioner according to the position deviation of the central position of the rectangular target frame corresponding to the human body target relative to the central point of the image; and obtaining the distance between the human target to be detected and the air conditioner according to the distance from the lower edge line of the rectangular target frame corresponding to the human target to the bottom of the image, the mapping relation between the distance from the lower edge line of the rectangular target frame to the bottom of the image and the distance from the human body to the air conditioner, which is obtained in advance. The invention can accurately acquire the direction and distance information of the human body and assist the air conditioner to make correct judgment.
Description
Technical Field
The invention relates to the field of intelligent home furnishing, in particular to a human body detection and positioning method and device applied to an air conditioner and an intelligent sensing system.
Background
The intelligent air conditioner is an important component of an intelligent home system, and mainly aims to realize intelligent management and control of the air conditioner, so that the optimal experience is obtained and the power resource is saved. At present, the intelligent air conditioner is still only in a concept stage, and the core of the intelligent air conditioner is a front-end sensing system and a rear-end motor control system.
The existing sensor module applied to the air conditioner generally takes a visible light camera sensor as a main part, and a few thermopile modules are used. The visible light module collects scene data, determines the position and the posture of a human body through a system sensing algorithm, assists the air conditioner motor control system to adjust the wind direction of wind power, and adjusts the temperature of the air conditioner by combining a temperature sensor arranged in the air conditioner. The thermopile sensor roughly determines the orientation and temperature of a human body through sensing scene heat distribution, and assists in controlling the adjustment of an air conditioner fan and a temperature sensing system.
The visible light camera module is not good at the formation of image under the not enough environment of light, greatly influences perception algorithm's detection effect. And the visible light camera module catches the texture detail of target better, has the risk of exposing the privacy, so its application on the air conditioner receives very big restriction. The temperature sensors can only measure local temperature, a large number of temperature sensors are needed for measuring the temperature of each area of the whole room, the cost is high, the error is large, the installation is inconvenient, and the popularization and the use of products are not facilitated.
The thermopile sensor module can capture a target and measure temperature at the same time, but the thermopile sensor module has small area array and weak energy, can only detect a limited distance range, adopts the existing human body detection and positioning method based on the thermopile sensor, has poor target detection and temperature measurement effects outside 2-3 meters, and provides insufficient criteria to assist an air conditioner to make correct judgment.
Disclosure of Invention
In order to solve at least part of the problems in the prior art, the invention provides a human body detection and positioning method and device applied to an air conditioner and an intelligent sensing system.
The invention is realized by the following steps:
in a first aspect, the present invention provides a human body detection and positioning method applied to an air conditioner, comprising the following steps:
carrying out threshold segmentation on a human body infrared image acquired in an air-conditioning scene;
extracting and tracking target edges of the image subjected to threshold segmentation to obtain a plurality of complete contours;
carrying out circumscribed rectangle description on each complete outline to form a plurality of rectangular target frames;
screening the plurality of rectangular target frames through human body constraints to screen out rectangular target frames corresponding to human body targets;
determining the orientation of the human body target to be detected relative to the air conditioner according to the position deviation of the central position of the rectangular target frame corresponding to the human body target relative to the central point of the image;
and obtaining the distance between the human target to be detected and the air conditioner according to the distance from the lower edge line of the rectangular target frame corresponding to the human target to the bottom of the image, the mapping relation between the distance from the lower edge line of the rectangular target frame to the bottom of the image and the distance from the human body to the air conditioner, which is obtained in advance.
Further, the threshold segmentation of the human infrared image acquired in the air-conditioning scene specifically includes:
traversing the whole image by using a horizontal template with the width of L and the step length of 1, calculating the gray value mean value of the covered pixels of the horizontal template when traversing to a position, taking the calculated gray value mean value as a low threshold value, adding a preset offset to the low threshold value as a high threshold value, thresholding each current pixel according to the following formulas (1), (2) and (3) until the threshold segmentation of the whole image is completed, and stopping traversing;
TH(x，y)＝TL(x，y)+offset (3)
wherein, B (x, y) is the gray value of the pixel point (x, y) after threshold segmentation, I (x, y) is the original gray value of the pixel point (x, y), TL(x, y) is a low threshold, TH(x, y) is a high threshold and offset is an offset.
Further, the method further includes performing hole filling and denoising processing on the image after threshold segmentation, and specifically includes:
performing opening operation by using a 3x3 corrosion template and an expansion template to remove fine noise;
the noise was filled in with a closed operation using a 10x1 expansion template and a 3x3 erosion template.
Further, the extracting and tracking the target edge of the infrared image after the threshold segmentation to obtain a plurality of complete contours specifically includes:
judging whether the binary image after threshold segmentation is a 0-1 image, and if not, performing 0-1 mapping conversion to obtain a 0-1 image;
the 4-neighborhood or 8-neighborhood chain code is used for tracking boundary pixels, one pixel on the target boundary of the 0-1 image is firstly found, then the other pixels on the target boundary are sequentially found by taking the pixel as a starting point until the starting point is reached, the tracking of the target boundary is completed, and the tracking of all the target boundaries is sequentially completed by adopting the method.
Further, screening a plurality of rectangular target frames through human body constraint to screen out the rectangular target frame corresponding to the human body target, specifically including:
and screening the plurality of rectangular target frames by adopting at least two human body constraints of human body aspect ratio constraint, human body maximum and minimum area constraint and human body heel landing constraint to screen out the rectangular target frames corresponding to the human body targets.
Further, the determining the orientation of the human target to be detected relative to the air conditioner according to the position deviation of the central position of the rectangular target frame corresponding to the human target relative to the central point of the image specifically includes:
the coordinate of the central point of the image is (x)0，y0) The coordinates of the upper left corner and the lower right corner of the rectangular target frame corresponding to the human body target are respectively (x)min，ymin) And (x)max，ymax) And the offset angle of the human body relative to the air conditioner air outlet is as follows:
where θ is the offset angle, (x)center，ycenter) Is the coordinate of the center position of the human body.
In a second aspect, the present invention provides a human body detecting and positioning device for an air conditioner, comprising:
the threshold segmentation module is used for carrying out threshold segmentation on the human body infrared image acquired in the air-conditioning scene;
the contour extraction module is used for extracting and tracking target edges of the image subjected to threshold segmentation to obtain a plurality of complete contours;
the external rectangle description module is used for carrying out external rectangle description on each complete outline to form a plurality of rectangular target frames;
the target frame screening module is used for screening the plurality of rectangular target frames through human body constraints to screen out rectangular target frames corresponding to human body targets;
the target position determining module is used for determining the position of the human body target to be detected relative to the air conditioner according to the position deviation of the central position of the rectangular target frame corresponding to the human body target relative to the central point of the image;
and the target distance determining module is used for obtaining the distance between the human target to be detected and the air conditioner according to the distance between the lower edge line of the rectangular target frame corresponding to the human target and the bottom of the image, the mapping relation between the distance between the lower edge line of the rectangular target frame and the bottom of the image and the distance between the human body and the air conditioner, and the mapping relation is obtained in advance.
In a third aspect, the invention provides an intelligent sensing system applied to an air conditioner, which comprises an image acquisition unit, an image analysis unit, an information transmission unit and an air conditioner control center, wherein the image acquisition unit is used for acquiring images;
the image acquisition unit comprises an image acquisition module, and the image acquisition module is used for acquiring a human body image sequence under an air-conditioning scene through a thermal infrared imager;
the image analysis unit comprises a human body detection and positioning module and a human body temperature measurement module, wherein the human body detection and positioning module is used for detecting and positioning a human body by adopting any one of the methods; the human body temperature measurement module is used for detecting the human body temperature and the scene temperature;
the information transmission unit packs the information acquired by the image analysis unit and sends the information to the air conditioner control center;
and the air conditioner control center is used for adjusting the air speed, the steering and the temperature setting of the air conditioner through the air conditioner control unit after multi-data fusion analysis according to the acquired information.
Furthermore, the image acquisition unit further comprises a distortion correction module for performing distortion correction on the image acquired by the image acquisition module.
Further, the image analysis unit further comprises a kicking detection module for detecting the kicking state of the human body in the sleeping state.
Compared with the prior art, the invention has the following beneficial effects:
according to the human body detection and positioning method and device applied to the air conditioner and the intelligent sensing system, threshold segmentation, contour extraction, external rectangle description, target frame screening, target direction determination and target distance determination are sequentially carried out on the basis of the human body infrared image, so that the direction and distance information of the human body can be accurately acquired, and the air conditioner is assisted to make correct judgment; this be applied to intelligent perception system of air conditioner neither receives the light restriction, can carry out scene region temperature measurement again, does not infringe the privacy, and one set of perfect scheme realizes human detection and location, temperature measurement and plays and is detected the function, low cost, and the algorithm complexity is low, and the integrated embedded equipment of being convenient for uses, the development of the intelligent air conditioner trade of very big boosting.
Drawings
Fig. 1 is a flowchart of a human body detection and positioning method applied to an air conditioner according to an embodiment of the present invention;
FIG. 2 is an infrared image collected in an indoor environment according to an embodiment of the present invention;
fig. 3 is a binary image after threshold segmentation according to an embodiment of the present invention;
FIG. 4 is an image after hole filling and denoising processing according to an embodiment of the present invention;
FIG. 5 is an image of an object after extraction according to an embodiment of the present invention;
FIG. 6 is an image after the circumscribed rectangle is described according to the embodiment of the present invention;
FIG. 7 is a final acquired human target image provided by an embodiment of the present invention;
fig. 8 is a block diagram of a human body detecting and positioning device applied to an air conditioner according to an embodiment of the present invention;
fig. 9 is a block diagram of an intelligent sensing system applied to an air conditioner according to an embodiment of the present invention;
fig. 10 is a schematic diagram of a specific operation process of the image analysis unit according to the embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, an embodiment of the present invention provides a human body detection and positioning method applied to an air conditioner, which is based on a human body infrared image collected in an air-conditioning scene. As shown in fig. 2, the infrared image data collected in an indoor environment is subjected to non-uniform correction and dynamic range compression before being output, and the imaging effect does not affect the detection result, so that image processing is not required to improve the imaging quality. The method comprises the following steps:
s1, carrying out threshold segmentation on the human body infrared image collected in the air-conditioning scene;
traversing the whole image by using a horizontal template with the width of L and the step length of 1, calculating the gray value mean value of the covered pixels of the horizontal template when traversing to a position, taking the calculated gray value mean value as a low threshold value, adding a preset offset to the low threshold value as a high threshold value, thresholding each current pixel according to the following formulas (1), (2) and (3) until the threshold segmentation of the whole image is completed, and stopping traversing;
TH(x，y)＝TL(x，y)+offset (3)
wherein, B (x, y) is the gray value of the pixel point (x, y) after threshold segmentation, I (x, y) is the original gray value of the pixel point (x, y), TL(x, y) is a low threshold, TH(x, y) is a high threshold and offset is an offset.
The binary image after threshold segmentation is shown in fig. 3. The template width is related to the width of a step signal formed by a human body and a surrounding background, and can be adjusted according to a specific scene.
S2, carrying out cavity filling and denoising processing on the image after threshold segmentation;
the divided binary image has a lot of noise and holes, so that noise removal and hole filling processing are required. The functions can be realized by morphological open and close operation, the boundary can be smoothed, and the processing effect is good. In this embodiment, the following method is adopted to perform hole filling and denoising processing on the image after threshold segmentation:
performing opening operation by using a 3x3 corrosion template and an expansion template to remove fine noise;
the noise was filled in with a closed operation using a 10x1 expansion template and a 3x3 erosion template.
The image after the hole filling and denoising process is shown in fig. 4.
This step S2 may not be performed when the image forming effect is good.
S3, extracting and tracking the target edge of the infrared image subjected to cavity filling and denoising processing to obtain a plurality of complete contours;
firstly, judging whether a binary image after threshold segmentation is a 0-1 image, and if not, performing 0-1 mapping conversion to the 0-1 image;
then tracking boundary pixels by using a 4-neighborhood or 8-neighborhood chain code, firstly finding a pixel on a 0-1 image target boundary, then sequentially finding other pixels on the target boundary by taking the pixel as a starting point until the starting point is reached, completing the tracking of the target boundary, and sequentially completing the tracking of all target boundaries by adopting the method.
Specifically, the boundary bounding relation determining step is as follows:
(1) the binary image is traversed until the finding of the starting point of the outer boundary or the hole boundary stops (the outer boundary starting point is considered if the point satisfies both the conditions of the outer boundary starting point and the hole boundary starting point), assuming that the starting point is (i, j). The newly found boundary is labeled with a particular label, which is called the boundary consecutive Number (NBD). The boundary continuation number starts from 2 because 1 has been assigned to the boundary start point.
(2) The new boundary to be found is traced from the start of the boundary, marking the pixel values on the boundary as consecutive numbers of the boundary.
(3) If the current boundary (not the entire boundary but one of the segments) is formed between the 0-component (hole) and the 1-component, the pixel value of the current boundary pixel point (x, y) is changed to-NBD (negative number of the number value), otherwise it is set to NBD. This prevents pixel (x, y) from becoming the starting point for an existing boundary.
(4) And after the complete boundary is tracked, the position of the last traversal is obtained again until the lower right corner of the image is reached, the traversal is stopped, and the algorithm is ended.
When the boundary pixels are tracked by using a 4-neighborhood or 8-neighborhood chain code, assuming that an input image is F ═ { F (i, j) }, firstly setting frame pixels as mark number starting sequence numbers 1 to form a special hole boundary, starting to traverse the image until F (i, j)) ≠ 0 with the step length of 1, and setting the sequence number LNBD of a parent boundary to 1 when traversing to a new row every time. The new row is equivalent to that the parent boundaries meeting the new boundary are all frames, so the frames are set to be 1;
the specific tracking strategy is as follows:
(1) f (i, j) ≠ 0, which of the following three cases is determined:
(a) if f (i, j) is 1 and f (i, j-1) is 0 then (i, j) is the starting point for the outer boundary to be tracked, NBD + +, (i, j-1) → (i2, j 2);
(b) if f (i, j) ≧ 1 and f (i, j +1) ≧ 0, (i, j) is the starting point to track the hole boundary, NBD + +, (i, j +1) → (i2, j 2); if f (i, j) > 1, f (i, j) → LNBD;
(c) otherwise, entering the step (4).
(2) Determining the parent boundary of the tracking boundary according to the type of the newly found boundary and the parent boundary mark serial number LNBD;
(3) tracking the found boundary from the starting point (i, j), specifically as the substeps (a) to (e):
(a) (ii) finding non-0 pixels from its 8-neighborhood starting from (i2, j2) clockwise around (i, j), assigning the first pixel coordinate not being 0 as (i1, j1), and if there are no non-0 pixels in the neighborhood, assigning the value of f (i, j) as-NBD, proceeding to step (4);
(b)(i1，j1)→(i2，j2)，(i，j)→(i3，j3)；
(c) starting from the next pixel of (i2, j2) in the counter-clockwise direction, the first non-0 pixel is checked counter-clockwise around the current pixel (i3, j3), changing its coordinate to (i4, j 4);
(d) changing the pixel value f (i3, j3) of the pixel (i3, j3) as follows;
(ii) if (i3, j3+1) is a 0 pixel value point, then f (i3, j3) is-NBD;
② if (i3, j3+1) is not a 0 pixel value point and f (i3, j3) is 1, then f (i3, j3) is NBD;
③ otherwise, not changing the value of the original f (i3, j 3);
(e) if (i4, j4) is equal to (i, j) and (i3, j3) is equal to (i1, j1), returning to the starting point, and entering the step (4); otherwise, (i3, j3) → (i2, j2), (i4, j4) → (i3, j3), returning to step (c);
(4) if f (i, j) ≠ 1, then | f (i, j) | → LNBD, starting traversal from the pixel point (i, j +1) until the lower right corner of the image ends the algorithm.
The image after target edge extraction and tracking is shown in fig. 5, and the boundary points have been stored in the corresponding containers.
S4, carrying out circumscribed rectangle description on each complete outline to form a plurality of rectangular target frames;
specifically, the maximum value and the minimum value of x and y in coordinates (x, y) of each contour point of the complete contour are found, two vertexes of the upper left corner and the lower right corner of the circumscribed rectangle are determined, the vertexes and the width and the height of the circumscribed rectangle are further determined, the circumscribed rectangle description is performed on each complete contour, and a plurality of rectangular target frames are formed, as shown in fig. 6.
S5, screening the plurality of rectangular target frames through human body constraint, screening out the rectangular target frames corresponding to the human body targets, and extracting parameter information of the screened target frames;
fig. 6 shows that some interfering heat sources are also detected, which requires target frame screening, and these interfering heat sources are removed to screen out the rectangular target frame corresponding to the human body target. Specifically, at least two human body constraints of human body aspect ratio constraint, human body maximum and minimum area constraint and human body heel landing constraint are adopted to screen a plurality of rectangular target frames, and the rectangular target frames corresponding to the human body targets are screened out.
The method for screening a plurality of rectangular target frames by adopting each human body constraint comprises the following steps:
(1) aspect ratio constraint
Through a large number of experimental statistics, the aspect ratio of the pedestrian target frame is almost all between 0.25 and 0.5. The method comprises the following steps of screening a plurality of rectangular target frames through the following constraint conditions to screen out the rectangular target frames meeting requirements:
0.25≤width/height≤0.5
width and height are the width and height of the rectangular target frame, respectively.
(2) Maximum and minimum area constraints
Minimum and maximum area thresholds can be determined from the image resolution size, the experimental sample image size is 384x288, the selected minimum area threshold 2000, the maximum area threshold 12000. The method comprises the following steps of screening a plurality of rectangular target frames through the following constraint conditions to screen out the rectangular target frames meeting requirements:
2000≤width*height≤12000
width and height are the width and height of the rectangular target frame, respectively.
(3) Heel strike restraint
And screening the rectangular target frame by utilizing the physical constraint that the heel of the human body stands on the ground. Specifically, the positions of the ground at the whole image at different distances from the camera can be marked in advance, the distance from the human body to the camera is preliminarily deduced through the size of the rectangular target frame, the position of the lower edge of the rectangular target frame is compared with the position of the ground mark at the distance, the heel contact ground is judged if the error is within a certain range, otherwise, the heel is judged not to be in contact with the ground, and the rectangular target frame meeting the heel contact ground condition is screened out.
Through the screening strategy, some rectangular target frames corresponding to the interference heat sources can be excluded, only the rectangular target frames corresponding to the human body target are left, and the image obtained after screening the rectangular target frames is shown in fig. 7.
S6, determining the orientation of the human body target to be detected relative to the air conditioner according to the position deviation of the central position of the rectangular target frame corresponding to the human body target relative to the central point of the image;
after a rectangular area of the human body target is obtained, the target position is easily determined according to the center position of the rectangle. Suppose the coordinates of the center point of the image are (x)0，y0) The coordinates of the upper left corner and the lower right corner of the rectangular target frame are respectively (x)min，ymin) And (x)max，ymax) Then, the deviation angle of the human body relative to the air conditioner air outlet can be approximately expressed as:
where θ is the offset angle, (x)center，ycenter) Is the coordinate of the center position of the human body.
And S7, obtaining the distance between the human target to be detected and the air conditioner according to the distance between the lower edge line of the rectangular target frame corresponding to the human target and the bottom of the image, the mapping relation between the distance between the lower edge line of the rectangular target frame and the bottom of the image and the distance between the human body and the air conditioner, which is obtained in advance.
Because the same human body has different postures (such as the front and the side faces are opposite to the thermal imager), the imaging widths are different and the heights are the same. Different human targets have different heights, sizes and rectangular target frames with different heights and areas. Therefore, the mapping of the distance from the human body to the air conditioner by using the width, the height and the area of the rectangular target frame as calibration items is inaccurate. Therefore, the invention provides the distance from the heel, namely the lower edge line of the rectangular target frame, to the bottom of the image as a calibration item.
Specifically, the mapping relation between the distance from the lower edge line of the rectangular target frame to the bottom of the image and the distance from the human body to the air conditioner is obtained in advance through sampling. Taking the distance from the lower line of the rectangular target frame to the bottom of the image as a calibration item, and performing non-equal step length calibration according to a perspective principle, wherein the closer the distance is, the more effective pixels are, sparse sampling can be performed; the farther away the distance is, the unit distance of a single pixel is large, dense sampling can be performed, and finally, an image distance-real distance mapping table is formed and used as a standard for determining the distance, wherein the image distance refers to the distance from the lower edge line of the rectangular target frame to the bottom of the image, and the real distance refers to the distance from the human body to the air conditioner. And finally, searching an image distance-real distance mapping table according to the distance from the lower edge line of the rectangular target frame to the bottom of the image, wherein the searched real distance result is the distance from the human body to the air conditioner.
Based on the same inventive concept, the embodiment of the invention also provides a human body detection and positioning device applied to an air conditioner, and as the principle of the device for solving the technical problem is similar to that of the method embodiment, the implementation of the device can refer to the implementation of the method, and repeated parts are not repeated.
As shown in fig. 8, a human body detecting and positioning device applied to an air conditioner according to an embodiment of the present invention is used for implementing the method of the above embodiment, and the device includes
The threshold segmentation module 11 is used for performing threshold segmentation on the human body infrared image acquired in the air-conditioning scene;
the contour extraction module 13 is configured to perform target edge extraction and tracking on the image after threshold segmentation to obtain a plurality of complete contours;
the circumscribed rectangle description module 14 is used for performing circumscribed rectangle description on each complete outline to form a plurality of rectangular target frames;
the target frame screening module 15 is used for screening the plurality of rectangular target frames through human body constraints to screen out rectangular target frames corresponding to human body targets;
the target position determining module 16 is configured to determine a position of the human target to be detected relative to the air conditioner according to a position deviation of a center position of the rectangular target frame corresponding to the human target relative to a center point of the image;
and the target distance determining module 17 is configured to obtain the distance between the human target to be detected and the air conditioner according to the distance between the lower edge line of the rectangular target frame corresponding to the human target and the bottom of the image, and the mapping relationship between the distance between the lower edge line of the rectangular target frame and the bottom of the image and the distance between the human body and the air conditioner, which are obtained in advance.
The preferred embodiment further includes a cavity filling and denoising processing module 12, configured to perform cavity filling and denoising processing on the image after the threshold segmentation.
As shown in fig. 9, an embodiment of the present invention further provides an intelligent sensing system applied to an air conditioner, including an image acquisition unit, an image analysis unit, an information transmission unit, and an air conditioner control center;
the image acquisition unit comprises an image acquisition module, the image acquisition module comprises a thermal infrared imager arranged below the air conditioner and is used for acquiring a human body image sequence in an air-conditioning scene; in order to acquire wider scene information, a wide-angle lens is adopted, and a collected image needs to be subjected to distortion correction and then provided to an image analysis unit for further processing. Preferably, the image acquisition unit further comprises a distortion correction module for performing distortion correction on the image acquired by the image acquisition module.
The image analysis unit is used for processing and analyzing the acquired image data and extracting the human body direction information and the temperature information. The air conditioner body temperature measuring device comprises a human body detecting and positioning module and a human body temperature measuring module, wherein the human body detecting and positioning module is used for detecting and positioning a human body by adopting the human body detecting and positioning method applied to the air conditioner in the embodiment; the human body temperature measurement module is used for detecting the human body temperature and the scene temperature. Preferably, the image analysis unit further comprises a kicking detection module for detecting a kicking state of the human body in the sleep state.
The three functional modules of the image analysis unit are mutually independent and can be manually set to be closed or opened. And when the human body detection and positioning function is closed, the human body temperature measurement function is started, and the highest temperature and the lowest temperature of the scene and the position information of the scene are detected. The kick detection function can specify which time period to turn on by setting the time.
After receiving the image data processed by the image acquisition unit, detecting whether the human body detection and positioning functional module is started, if the human body detection and positioning functional module is started, carrying out image segmentation by an analysis algorithm to extract foreground targets, and counting the number of the targets and the position of the positioning target, wherein the target number and the position comprise the distance and the direction. And storing the calculation result into a temporary data container to wait for encoding and packaging. And if the human body detection and positioning functions are not started, directly setting 0 in the corresponding storage area.
And detecting whether the human body temperature measurement function is started, checking whether the human body detection and positioning function is started if the human body temperature measurement function is started, acquiring the detected human body position information if the human body temperature measurement and positioning function is started, and acquiring the head temperature and the highest temperature and the lowest temperature of each scene of the human body. And if the human body detection function is not started, mapping a temperature measurement curve according to the image AD value to acquire the highest temperature and the lowest temperature of the scene and the position information of the highest temperature and the lowest temperature. If the temperature measuring function is not started, directly setting 0 in the temperature measuring result container.
And detecting whether the kicked quilt detection function is started, if so, positioning the position of the human body and the position of the quilt by a kicked quilt detection algorithm, calculating the coverage rate of the quilt on the human body, and obtaining the temperature of the head of the human body. The results are saved to the corresponding container. If the kick-detected function is not turned on, 0 is directly set in the corresponding storage area.
The specific operation of the image analysis unit is shown in fig. 10.
The information transmission unit packs the human body number, direction, temperature and quilt state information acquired by the image analysis unit and sends the information to the air conditioner control center;
and the air conditioner control center is used for adjusting the air conditioner wind speed, steering and temperature setting through the air conditioner control unit after multi-data fusion analysis according to the acquired information.
In summary, the human body detection and positioning method, the human body detection and positioning device and the intelligent sensing system applied to the air conditioner provided by the embodiments of the present invention sequentially perform threshold segmentation, cavity filling and denoising processing, contour extraction, external rectangle description, target frame screening, target orientation determination and target distance determination based on the human body infrared image, and can more accurately acquire orientation and distance information of a human body to assist the air conditioner in making a correct judgment; this be applied to intelligent perception system of air conditioner neither receives the light restriction, can carry out scene region temperature measurement again, does not infringe the privacy, and one set of perfect scheme realizes human detection and location, temperature measurement and plays and is detected the function, low cost, and the algorithm complexity is low, and the integrated embedded equipment of being convenient for uses, the development of the intelligent air conditioner trade of very big boosting.
Those of ordinary skill in the art will appreciate that all or part of the steps of the various methods of the embodiments may be implemented by associated hardware as instructed by a program, which may be stored on a computer-readable storage medium, which may include: read Only Memory (ROM), Random Access Memory (RAM), magnetic or optical disks, and the like.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (10)
1. A human body detection and positioning method applied to an air conditioner is characterized by comprising the following steps:
carrying out threshold segmentation on a human body infrared image acquired in an air-conditioning scene;
extracting and tracking target edges of the image subjected to threshold segmentation to obtain a plurality of complete contours;
carrying out circumscribed rectangle description on each complete outline to form a plurality of rectangular target frames;
screening the plurality of rectangular target frames through human body constraints to screen out rectangular target frames corresponding to human body targets;
determining the orientation of the human body target to be detected relative to the air conditioner according to the position deviation of the central position of the rectangular target frame corresponding to the human body target relative to the central point of the image;
and obtaining the distance between the human target to be detected and the air conditioner according to the distance from the lower edge line of the rectangular target frame corresponding to the human target to the bottom of the image, the mapping relation between the distance from the lower edge line of the rectangular target frame to the bottom of the image and the distance from the human body to the air conditioner, which is obtained in advance.
2. The human body detection and positioning method applied to the air conditioner as claimed in claim 1, wherein the threshold segmentation of the human body infrared image collected under the air conditioning scene specifically comprises:
traversing the whole image by using a horizontal template with the width of L and the step length of 1, calculating the gray value mean value of the covered pixels of the horizontal template when traversing to a position, taking the calculated gray value mean value as a low threshold value, adding a preset offset to the low threshold value as a high threshold value, thresholding each current pixel according to the following formulas (1), (2) and (3) until the threshold segmentation of the whole image is completed, and stopping traversing;
TH(x，y)＝TL(x，y)+offset (3)
wherein, B (x, y) is the gray value of the pixel point (x, y) after threshold segmentation, I (x, y) is the original gray value of the pixel point (x, y), TL(x, y) is a low threshold, TH(x, y) is a high threshold and offset is an offset.
3. The method for detecting and locating the human body applied to the air conditioner as claimed in claim 1, further comprising performing a hole filling and denoising process on the image after the threshold segmentation, specifically comprising:
performing opening operation by using a 3x3 corrosion template and an expansion template to remove fine noise;
the noise was filled in with a closed operation using a 10x1 expansion template and a 3x3 erosion template.
4. The human body detection and positioning method applied to the air conditioner as claimed in claim 1, wherein the extracting and tracking of the target edge is performed on the infrared image after the threshold segmentation to obtain a plurality of complete contours, specifically comprising:
judging whether the binary image after threshold segmentation is a 0-1 image, and if not, performing 0-1 mapping conversion to obtain a 0-1 image;
the 4-neighborhood or 8-neighborhood chain code is used for tracking boundary pixels, one pixel on the target boundary of the 0-1 image is firstly found, then the other pixels on the target boundary are sequentially found by taking the pixel as a starting point until the starting point is reached, the tracking of the target boundary is completed, and the tracking of all the target boundaries is sequentially completed by adopting the method.
5. The human body detection and positioning method applied to the air conditioner as claimed in claim 1, wherein the screening of the plurality of rectangular target frames through human body constraints to screen out the rectangular target frames corresponding to the human body targets specifically comprises:
and screening the plurality of rectangular target frames by adopting at least two human body constraints of human body aspect ratio constraint, human body maximum and minimum area constraint and human body heel landing constraint to screen out the rectangular target frames corresponding to the human body targets.
6. The human body detection and positioning method applied to the air conditioner as claimed in claim 1, wherein the determining the orientation of the human body target to be detected with respect to the air conditioner according to the position deviation of the center position of the rectangular target frame corresponding to the human body target with respect to the image center point specifically comprises:
the coordinate of the central point of the image is (x)0，y0) The coordinates of the upper left corner and the lower right corner of the rectangular target frame corresponding to the human body target are respectively (x)min，ymin) And (x)max，ymax) And the offset angle of the human body relative to the air conditioner air outlet is as follows:
where θ is the offset angle, (x)center，ycenter) Is the coordinate of the center position of the human body.
7. A human body detection and positioning device applied to an air conditioner is characterized by comprising:
the threshold segmentation module is used for carrying out threshold segmentation on the human body infrared image acquired in the air-conditioning scene;
the contour extraction module is used for extracting and tracking target edges of the image subjected to threshold segmentation to obtain a plurality of complete contours;
the external rectangle description module is used for carrying out external rectangle description on each complete outline to form a plurality of rectangular target frames;
the target frame screening module is used for screening the plurality of rectangular target frames through human body constraints to screen out rectangular target frames corresponding to human body targets;
the target position determining module is used for determining the position of the human body target to be detected relative to the air conditioner according to the position deviation of the central position of the rectangular target frame corresponding to the human body target relative to the central point of the image;
and the target distance determining module is used for obtaining the distance between the human target to be detected and the air conditioner according to the distance between the lower edge line of the rectangular target frame corresponding to the human target and the bottom of the image, the mapping relation between the distance between the lower edge line of the rectangular target frame and the bottom of the image and the distance between the human body and the air conditioner, and the mapping relation is obtained in advance.
8. The utility model provides an intelligent perception system for air conditioner which characterized in that: the system comprises an image acquisition unit, an image analysis unit, an information transmission unit and an air conditioner control center;
the image acquisition unit comprises an image acquisition module, and the image acquisition module is used for acquiring a human body image sequence under an air-conditioning scene through a thermal infrared imager;
the image analysis unit comprises a human body detection and positioning module and a human body temperature measurement module, wherein the human body detection and positioning module is used for detecting and positioning a human body by adopting the method of any one of claims 1 to 6; the human body temperature measurement module is used for detecting the human body temperature and the scene temperature;
the information transmission unit packs the information acquired by the image analysis unit and sends the information to the air conditioner control center;
and the air conditioner control center is used for adjusting the air speed, the steering and the temperature setting of the air conditioner through the air conditioner control unit after multi-data fusion analysis according to the acquired information.
9. The intelligent sensing system applied to the air conditioner as claimed in claim 8, wherein: the image acquisition unit further comprises a distortion correction module for performing distortion correction on the image acquired by the image acquisition module.
10. The intelligent sensing system applied to the air conditioner as claimed in claim 8, wherein: the image analysis unit further comprises a kicking detection module for detecting the kicking state of the human body in the sleeping state.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111199532.9A CN114120354A (en) | 2021-10-14 | 2021-10-14 | Human body detection and positioning method and device applied to air conditioner and intelligent sensing system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111199532.9A CN114120354A (en) | 2021-10-14 | 2021-10-14 | Human body detection and positioning method and device applied to air conditioner and intelligent sensing system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114120354A true CN114120354A (en) | 2022-03-01 |
Family
ID=80375924
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111199532.9A Pending CN114120354A (en) | 2021-10-14 | 2021-10-14 | Human body detection and positioning method and device applied to air conditioner and intelligent sensing system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114120354A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114636235A (en) * | 2022-03-07 | 2022-06-17 | 青岛海信日立空调系统有限公司 | Air conditioner and control method thereof |
| CN115823645A (en) * | 2022-12-23 | 2023-03-21 | 西南林业大学 | Electric heater safety control system and method based on recent human body identification |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107657639A (en) * | 2017-08-09 | 2018-02-02 | 武汉高德智感科技有限公司 | A kind of method and apparatus of quickly positioning target |
| CN110084177A (en) * | 2019-04-24 | 2019-08-02 | 珠海格力电器股份有限公司 | Positioning system, method, control system, air conditioner and storage medium |
-
2021
- 2021-10-14 CN CN202111199532.9A patent/CN114120354A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107657639A (en) * | 2017-08-09 | 2018-02-02 | 武汉高德智感科技有限公司 | A kind of method and apparatus of quickly positioning target |
| CN110084177A (en) * | 2019-04-24 | 2019-08-02 | 珠海格力电器股份有限公司 | Positioning system, method, control system, air conditioner and storage medium |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114636235A (en) * | 2022-03-07 | 2022-06-17 | 青岛海信日立空调系统有限公司 | Air conditioner and control method thereof |
| CN115823645A (en) * | 2022-12-23 | 2023-03-21 | 西南林业大学 | Electric heater safety control system and method based on recent human body identification |
| CN115823645B (en) * | 2022-12-23 | 2023-11-21 | 西南林业大学 | Electric heater safety control system and method based on nearest human body identification |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liu et al. | A detection and recognition system of pointer meters in substations based on computer vision | |
| CN109785337B (en) | In-column mammal counting method based on example segmentation algorithm | |
| CN111339951A (en) | Body temperature measuring method, device and system | |
| CN109685078B (en) | Infrared image identification method based on automatic annotation | |
| CN106778737B (en) | A kind of license plate antidote, device and a kind of video acquisition device | |
| CN108229475B (en) | Vehicle tracking method, system, computer device and readable storage medium | |
| CN111563469A (en) | Method and device for identifying irregular parking behaviors | |
| CN107657639A (en) | A kind of method and apparatus of quickly positioning target | |
| CN108470356B (en) | Target object rapid ranging method based on binocular vision | |
| WO2020252974A1 (en) | Method and device for tracking multiple target objects in motion state | |
| CN108197604A (en) | Fast face positioning and tracing method based on embedded device | |
| CN113160161B (en) | Method and device for detecting defects at edge of target | |
| CN113340909B (en) | Glue line defect detection method based on machine vision | |
| CN110189375A (en) | A kind of images steganalysis method based on monocular vision measurement | |
| CN110414385B (en) | Lane line detection method and system based on homography transformation and characteristic window | |
| CN114120354A (en) | Human body detection and positioning method and device applied to air conditioner and intelligent sensing system | |
| CN111126393A (en) | Vehicle appearance refitting judgment method and device, computer equipment and storage medium | |
| CN112085700B (en) | Automatic extraction method, system and medium for weld joint region in X-ray image | |
| CN103913149B (en) | A kind of binocular range-measurement system and distance-finding method thereof based on STM32 single-chip microcomputer | |
| CN107016348A (en) | With reference to the method for detecting human face of depth information, detection means and electronic installation | |
| CN114820474A (en) | Train wheel defect detection method based on three-dimensional information | |
| CN108510544B (en) | Light strip positioning method based on feature clustering | |
| WO2012037085A1 (en) | Active lighting for stereo reconstruction of edges | |
| CN114608522B (en) | Obstacle recognition and distance measurement method based on vision | |
| CN110851978B (en) | Camera position optimization method based on visibility |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220301 |