CN107514745A - A kind of method and system of intelligent air condition stereoscopic vision positioning - Google Patents

A kind of method and system of intelligent air condition stereoscopic vision positioning Download PDF

Info

Publication number
CN107514745A
CN107514745A CN201710656194.4A CN201710656194A CN107514745A CN 107514745 A CN107514745 A CN 107514745A CN 201710656194 A CN201710656194 A CN 201710656194A CN 107514745 A CN107514745 A CN 107514745A
Authority
CN
China
Prior art keywords
human body
wind
air
conditioning
target area
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
Application number
CN201710656194.4A
Other languages
Chinese (zh)
Inventor
喻海中
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Feixun Data Communication Technology Co Ltd
Original Assignee
Shanghai Feixun Data Communication Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shanghai Feixun Data Communication Technology Co Ltd filed Critical Shanghai Feixun Data Communication Technology Co Ltd
Priority to CN201710656194.4A priority Critical patent/CN107514745A/en
Publication of CN107514745A publication Critical patent/CN107514745A/en
Pending legal-status Critical Current

Links

Abstract

The invention discloses a kind of method of intelligent air condition stereoscopic vision positioning, to solve the problems, such as that the scanning wind direction of air-conditioning can only be adjusted by default air conditioning mode in the prior art.The present invention includes step:S1:Human body target is imaged, and human body target image is pre-processed and corrected;S2:According to the human body target image after pretreatment and correction, human body target image and human body target image background are split;S3:According to the human body target image after segmentation, determine target area and obtain the three-dimensional data of the target area;S4:According to the three-dimensional data of the target area and target area, determine that air-conditioning sweeps wind region;S5:Wind region is swept according to air-conditioning, wind angle is swept by calculating acquisition air-conditioning.The correlation technique of stereoscopic vision can be combined using the present invention, the relevant position region of moving human body is positioned and measured, reaches targeting, dynamic adjusts the effect of air-conditioning wind direction.

Description

A kind of method and system of intelligent air condition stereoscopic vision positioning
Technical field
The present invention relates to the method and system neck of air-conditioning technical field, more particularly to a kind of positioning of intelligent air condition stereoscopic vision Domain.
Background technology
Current air-conditioning mainly carries out wind direction, wind intensity, the regulation and control of humidity by artificially presetting specific pattern, it is impossible to The active state for intelligently tracking people enters Mobile state adjustment.
Thermal camera imaging technique or comparative maturity in the prior art, are mainly carried out by detecting ambient temperature Imaging, compared to common video camera imaging, can by thermal camera into picture and ambient background form obvious profile, and Into picture be internally formed smooth UNICOM region;This is very big for that need not have to the imaging of image progress characteristic point mark Effect.
Stereoscopic vision basic fundamental, predominantly binocular stereo vision basic fundamental in the prior art, by simulating the double of people Eye, binocular stereo vision video camera is after demarcation, can be to external object at two after getting the inside and outside parameter of video camera The position of imaging in video camera, by computation vision difference, utilize the inside and outside parameter and Conversion Matrix of Coordinate of video camera, meter The three-dimensional position of target object is calculated, technique of binocular stereoscopic vision has been widely used in automatic driving at present, calculates The association area such as machine assisted surgery, reverse-engineering, interaction entertainment, measurement.
And thermal camera imaging technique and binocular stereo vision basic fundamental sky is not attached to also in the prior art In conditioning technology field.
Publication No. 102889666A Chinese patent, disclose a kind of air conditioning control method and air-conditioning.The air-conditioning control Method processed comprises the steps:Infrared sensor is set in indoor apparatus of air conditioner;Control infrared sensor Scanning Detction region with Infrared ray in detection zone, and export scanning result;Judge detection zone whether where someone and people according to scanning result Test position;Air conditioning operating mode is adjusted according to judged result.The invention is automatic by the infrared sensor being arranged in air-conditioning The information of environment space people is detected, and air conditioning operating mode can be automatically adjusted according to testing result, realizes the intelligence fortune of air-conditioning Row control.
Although foregoing invention can scan the region where people, automatically adjust operational mode, can not reach intelligence, Targeting, the dynamic effect for adjusting wind direction.
In summary, in order to solve the problems, such as intelligently determine that air-conditioning sweeps wind region, it is necessary to design one in the prior art Kind infrared Binocular Stereo Vision System can be utilized to target, dynamically adjust air-conditioning come the scope of activities of people in dynamic monitoring room Sweep the intelligent air condition stereoscopic vision localization method and system in wind region.
The content of the invention
The purpose of the present invention is for defect present in prior art, there is provided a kind of intelligent air condition stereoscopic vision positioning side Method and system, can be intelligent, need not preset dynamically and artificially air-conditioning scan pattern and sweep wind angle.
In order to realize the above object the present invention uses following technical scheme:
A kind of method of intelligent air condition stereoscopic vision positioning, including step:
S1:Human body target is imaged, and human body target image is pre-processed and corrected;
S2:According to the human body target image after pretreatment and correction, by human body target image and human body target image background Split;
S3:According to the human body target image after segmentation, determine target area and obtain the three-dimensional data of the target area;
S4:According to the three-dimensional data of the target area and target area, determine that air-conditioning sweeps wind region;
S5:Wind region is swept according to air-conditioning, wind angle is swept by calculating acquisition air-conditioning.
Wherein, human body target being imaged by thermal camera, the quantity of the thermal camera is for 2 or multiple, It is described human body target image and human body target image background are subjected to segmentation to include step:
Calculate the average background figure of human body target image background;
Split by the gray scale difference value of human body target image and the average background figure.
Further, obtaining the three-dimensional data of the target area includes step:
The marginal point of target area is extracted;
The target area marginal point that different thermal cameras extract is matched and the three-dimensional of target area is calculated Data.
Further, the posture of the human body target is identified, according to the posture of the human body target and the human body mesh Air-conditioning, which is preset, corresponding to the three-dimensional data selection of mark posture sweeps wind region.
Wherein determine that air-conditioning sweeps wind region and specifically includes step:
Determined to sweep the quantity in wind region according to the quantity of target area;
The circumscribed rectangular body of the target area is calculated according to the three-dimensional data of each target area;
Determine that air-conditioning sweeps wind region according to the circumscribed rectangular body of each target area.
When the air-conditioning sweep wind region for it is multiple when, the multiple air-conditioning is swept into wind region merging technique into always sweeping wind region;
Sweep when wind region is always swept described in wind, when sweeping to multiple air-conditionings and sweeping wind region intermediate blank region, adjust this and sweep wind Speed is that default blank sweeps wind velocity.
A kind of intelligent air condition stereoscopic vision alignment system, including:
Image-forming module:For being imaged to human body target;
Image preprocessing and correction module, for human body target image to be pre-processed and corrected;
Target Segmentation module, for human body target image and human body target image background to be split;
Characteristic area determining module, for determining target area and obtaining the three-dimensional data of the target area;
Wind area determination module is swept, for determining that air-conditioning sweeps wind region;
Wind angle adjustment module is swept, wind angle is swept for calculating acquisition air-conditioning.
Wherein image preprocessing and correction module includes:
Image pre-processing module, for carrying out noise suppressed or enhancing and to human body target image to human body target image Corrected;
Image correction module, for carrying out human body target image parameter correction according to default uncalibrated image.
The characteristic area determining module includes:
Target area determining module, for determining target area;
Three-dimensional data acquisition module, for obtaining the three-dimensional data of target area by calculating.
Further, including:
Identification module, for identifying the posture of the human body target;
Selecting module, for the selection pair of the three-dimensional data of the posture according to the human body target and the human body target posture The default air-conditioning answered sweeps wind region.
The wind area determination module of sweeping includes:
Quantity determining module, for being determined to sweep the quantity in wind region according to the quantity of target area;
Area determination module, for calculating the external of the target area according to the three-dimensional data of each target area Cuboid, and determine that air-conditioning sweeps wind region according to the circumscribed rectangular body of each target area;
Further, in addition to:
Merging module, for when it is multiple that the air-conditioning, which sweeps wind region, the multiple air-conditioning to be swept into wind region merging technique shape Into always sweeping wind region;
Wind velocity module is swept in adjustment, for sweeping wind region according to the air-conditioning always swept in wind region and multiple air-conditionings sweep wind White space between region carries out sweeping wind velocity adjustment.
The beneficial effects of the invention are as follows:By using infrared binocular stereo vision alignment system, indoor human body target is detected Temperature be imaged, simulate the eyes of human body, using the image space in different cameras, pass through computation vision difference, it is real The scope of activities of people in existing dynamic monitoring room, targeting, the dynamic region for adjusting air-conditioning and sweeping wind.
Brief description of the drawings
Fig. 1 is a kind of flow chart of the method for intelligent air condition stereoscopic vision positioning;
Fig. 2 is a kind of another flow chart of the method for intelligent air condition stereoscopic vision positioning;
Fig. 3 is a kind of Stereo matching schematic diagram of intelligent air condition stereoscopic vision localization method;
Fig. 4 is a kind of three-dimensional data chart of the acquisition characteristic point of intelligent air condition stereoscopic vision localization method;
Fig. 5 is that a kind of air-conditioning of intelligent air condition stereoscopic vision localization method sweeps wind angle-determining figure;
Fig. 6 is a kind of structural principle block diagram of intelligent air condition stereoscopic vision alignment system;
Fig. 7 is a kind of another structure principle chart of intelligent air condition stereoscopic vision alignment system.
Embodiment
It is the specific embodiment of the present invention and with reference to accompanying drawing below, technical scheme is further described, But the present invention is not limited to these embodiments.
Embodiment one
A kind of method of intelligent air condition stereoscopic vision positioning is present embodiments provided, it is shown as shown in figures 1 and 3, including Step:
S1:Human body target is imaged, and human body target image is pre-processed and corrected;
S2:According to the human body target image after pretreatment and correction, by human body target image and human body target image background Split;
S3:According to the human body target image after segmentation, determine target area and obtain the three-dimensional data of the target area;
S4:According to the three-dimensional data of the target area and target area, determine that air-conditioning sweeps wind region;
S5:Wind region is swept according to air-conditioning, wind angle is swept by calculating acquisition air-conditioning.
Wherein human body target is imaged by thermal camera, the quantity of the thermal camera is for 2 or multiple.
Stereoscopic vision basic fundamental, what is utilized in of the invention is binocular stereo vision basic fundamental, in order to improve video camera Computational accuracy, can be imaged using multiple video cameras.Binocular stereo vision basic fundamental, it is double by simulating the eyes of people Item stereo vision video camera is after demarcation, can be to external object in two shootings after getting the inside and outside parameter of video camera The position of imaging in machine, by computation vision difference, using the inside and outside parameter and Conversion Matrix of Coordinate of video camera, calculate The three-dimensional position of target object, at present technique of binocular stereoscopic vision be widely used in automatic driving, computer aided manufacturing The association area such as assistant's art, reverse-engineering, interaction entertainment, measurement.
The imaging of thermal camera is imaged by detecting infrared source, why using infrared binocular tri-dimensional Feel system and unconventional optics Binocular Stereo Vision System, it is because the temperature of human body is constant at 37 DEG C or so, therefore in institute Into in image, target person can form obvious profile with ambient background, and smooth UNICOM region is internally formed in portrait, so as to The segmentation difficulty of people is greatly reduced, improves the precision of segmentation.And special characteristic point mark need not be carried out to human body.
Image preprocessing includes image denoising, enhancing, and image is carried out using the distortion parameter of camera lens the phase such as to correct Close operation.
The correction of binocular stereo image, it is the correction parameter according to camera calibration, by the relevant matches in two images Position correction facilitates follow-up Stereo matching, calculates the parallax of corresponding points on same baseline.
The basic schematic diagram of wherein Stereo matching is as shown in Figure 3.
Human body target image after pretreatment and correction is entered with human body target background by gray value or characteristic point Row is split, and human body target image and human body target image background are divided into non-overlapping copies, has the region of respective feature, from And human body target image can be subjected to target area division.
The determination of target area includes the determination of single goal target area and the determination in multiple target region:
The determination in single goal region:Single goal region includes single human body target-single area and more human body targets-single Two kinds of region situation:
A. single human body target-single area:One-man's body target i.e. in image, the single human body segmentation region is mesh Mark region;
B. more human body target-single areas:The region is it is also contemplated that two kinds of situations:
The first situation:Blocked and caused by human body, i.e., three-dimensional human body target is mapped in the image of two dimension, forms weight Folded region, in such cases, the result of two human body Target Segmentations is a connected region, and the region is target area.
Second of situation:Do not block, the result of segmentation is the region of multiple connections, but distance is non-between connected region Chang Jin, it is therefore desirable to merge region.In such a case, it is possible to by calculating the centre-to-centre spacing between connected region two-by-two From, if distance be less than default a certain threshold value, two regions can be merged, combined region is target area.
The determination in multiple target region
Multiple target region is that the result of human body target segmentation is the region of multiple connections, while the centre-to-centre spacing of each connected region From more than default a certain threshold value, then multiple cut zone are target area.
Target area marginal point is matched, and calculated, obtains the three-dimensional data of target area.
Wherein the quantity of target area determines to sweep the quantity in wind region, and the three-dimensional data for calculating the target area of acquisition is That sweeps wind region sweeps wind volume;Wind volume of sweeping according to wind region is swept utilizes three-dimensional coordinate, and accurate adjustment air-conditioning sweeps wind angle.
The present embodiment, using the imaging of infrared binocular camera, segmentation, matching, calculate and determine sweeping wind region and sweeping for air-conditioning The angle in wind region, sweep wind region so as to reach intelligence, dynamic, need not artificially preset air-conditioning and sweep wind angle, improve Consumer's Experience.
Embodiment two
A kind of method of intelligent air condition stereoscopic vision positioning is present embodiments provided, as shown in Figures 2 to 5, including step:
Thermal camera is demarcated;
Human body target is imaged, and human body target image is pre-processed and corrected;
Calculate the average background figure of human body target image background;
Split by the gray scale difference value of human body target image and the average background figure;
Determine target area;
The marginal point of target area is extracted;
The target area marginal point that different thermal cameras extract is matched and the three-dimensional of target area is calculated Data;
The posture of the human body target is identified, according to the posture of the human body target and the three-dimensional of the human body target posture Air-conditioning, which is preset, corresponding to data selection sweeps wind region.
Determined to sweep the quantity in wind region according to the quantity of target area;
The circumscribed rectangular body of the target area is calculated according to the three-dimensional data of each target area;
Determine that air-conditioning sweeps wind region according to the circumscribed rectangular body of each target area;
Determine air-conditioning sweeps wind angle.
The wherein purpose of thermal camera demarcation, the inside and outside parameter of thermal camera is mainly obtained, for follow-up figure As correction, the correction of image pair, three-dimensional mapping matrix etc. is obtained, its major parameter includes:The focal length of video camera, camera center are sat Cursor position, (present invention acquiescence illustrates 2 video cameras to two video cameras, in order to improve positioning precision, can increase the number of video camera Amount) distance between center, the elevation angle of video camera, the distortion parameter matrix, the transformed matrix of each reference frame etc. of camera lens.
At present, the calibration technique of video camera is highly developed, and also efficiency is also very high (as classical opens just for stated accuracy Friendly chessboard method).It is infrared to be carried out by detecting infrared source because what is utilized in the present invention is infrared Binocular Stereo Vision System Imaging, therefore in calibration process, it is impossible to using conventional table top plate, it is necessary to consider that temperature difference forms suitable calibration image, Such as it is contemplated that with hollow out chessboard be placed in the stable thermal source of stable and uniform (such as human body, human body temperature are scheduled on 37 degrees centigrades very much, The background of chessboard imaging can be served as) before, so as to reduce the difficulty and error of detection chessboard angle point, improve the essence of demarcation Degree.
Wherein camera calibration is disposably completed in installation process, it is not necessary to is demarcated every time.
After demarcation, infrared binocular camera is imaged to human body target, and human body target image is pre-processed and corrected, With embodiment one.
Human body target image specifically includes step with the segmentation of human body target image background:
Calculate the average background figure of human body target image background;
Split by the gray scale difference value of human body target image and the average background figure.
The human body temperature detected is constant at 37 DEG C or so, therefore human body target image and human body target in infrared image Image background separates substantially, and inside human body target image-region, texture is smooth, can generally form the inner area of closing Domain, avoid in optical imagery and separate difficulty caused by complex background interference, generally by the binary segmentation such as Da-Jin algorithm method just Preferable segmentation result can be obtained.Meanwhile people is the target of activity, background is static object, therefore can be by using red The front and rear multiple image of outer binocular camera input calculates average background figure, by current human target's image and average background The difference of figure splits human body target image and human body target image background, further the precision of increase segmentation.
By the human body target image after segmentation, target area is determined by connecting to calculate, such as embodiment one;
The marginal point of target area is extracted, the target area marginal point progress that different thermal cameras are extracted Match somebody with somebody and calculate the three-dimensional data of target area, it is specific as follows:
The general principle of binocular stereo vision is to simulate the eyes of people, allows the target object of same three-dimensional world in two eyes Vision offset in eyeball, i.e. parallax, by the processing of brain, inverse mapping is into three-dimensional coordinate, so as to allow human brain to form the depth of three-dimensional Degree perceives.Same object is exactly imaged by Stereo matching in two video cameras, calculates the vision deviation of same position, and formation regards Poor matrix.
Stereo matching in the present invention only needs to match target area marginal point, due to being the matching of marginal point, And it is less in the absence of blocking or blocking, therefore the precision matched is higher, due to being the matching of sparse matrix, therefore matching efficiency Can be very high.
It is assumed that a certain target area is represented with rhombus, it is assumed that wherein solid line, dotted line rhombus are respectively left and right Imaging in video camera, wherein corresponding a certain characteristic point is mutually should be A, A ', it is d that coordinate (x, y), (x ', y), which then match parallax, =| x '-x |, then the parallax matrix group obtained turns into (x, y, d), using this group of parallax matrix with, before after camera calibration by The confederate matrix Q (4x4 matrix) that camera interior and exterior parameter and transformation matrix of coordinates are formed, you can reflect parallax matrix inversion Three dimensions is mapped to, so as to get the three-dimensional data of characteristic point.
The posture of the human body target is identified, according to the posture of the human body target and the three-dimensional of the human body target posture Air-conditioning, which is preset, corresponding to data selection sweeps wind region.
Specifically, such as the people that sleeps of a people to lie low and side, the wind region area of sweeping of the people to lie low is its plane face Product, the people that side is slept only cannot sweep wind region area from his area of plane come consider it, only consider because sweeping wind The area of plane can cause its side can not blow to air-conditioner wind.
So the scheme in the present embodiment may determine that the sleeping position of people, according to sleeping position, carry out different sweeping wind strategy.
Assuming that the quantity of above-mentioned target area is n, then can determine that air-conditioning sweeps wind region as n, utilizes above-mentioned target The three-dimensional coordinate data of provincial characteristics point, you can the n circumscribed rectangular bodies (containing cube) for sweeping wind region are calculated, it is the n long Cube is that n positioned sweep wind region.
Air-conditioning is swept into wind function mark off n and sweep wind functional areas, map n Ge Saofeng areas one by one, so as to reach targeting, dynamic State, intelligently sweep wind function.
As shown in figure 5, when certain circumscribed rectangular body for sweeping wind region is ABCD-A1B1C1D1When, it may be determined that cuboid The three-dimensional coordinate point each pinpointed, according to the three-dimensional coordinate of air-conditioning position, you can calculate between any two points at air-conditioning Three-dimensional angle, wind angle is swept so as to accurately adjust air-conditioning.
When the air-conditioning sweep wind region for it is multiple when, the multiple air-conditioning is swept into wind region merging technique into always sweeping wind region;
Sweep when wind region is always swept described in wind, when sweeping to multiple air-conditionings and sweeping wind region intermediate blank region, adjust this and sweep wind Speed is that default blank sweeps wind velocity.
For example, determining have two to sweep wind region, a kind of scheme is to be integrated into one always to sweep wind region, carries out sweeping wind; Further scheme is, when sweeping wind, it sweeps the Origin And Destination of wind always to sweep wind region, and when sweeping wind, wind is swept when sweeping to two During the white space of the centre in region, the speed of wind angle is swept in adjustment.So that the efficiency and Consumer's Experience of intelligent air condition are entered One step is lifted.
The present embodiment, there is provided a kind of method of intelligent air condition stereoscopic vision positioning, demarcation from infrared binocular camera, Split by heat radiation images, increase the precision of segmentation;Only target area marginal point is matched so that matching precision Height, matching efficiency is high, and intelligently can enter Mobile state adjustment according to the three-dimensional data of human body target posture and human body target posture Sweep wind strategy;And air-conditioning swept into wind function mark off n and sweep wind functional areas, n air-conditioning Sao Feng area is mapped one by one, and can be according to sky Adjust the white space swept between the human body target in wind region and human body target region to carry out sweeping wind adjustment, so as to reach targeting, Dynamically, wind function is intelligently swept.
Embodiment three
A kind of intelligent air condition stereoscopic vision alignment system is present embodiments provided, as shown in figure 5, including:
Image-forming module:For being imaged to human body target;
Image preprocessing and correction module, for human body target image to be pre-processed and corrected;
Target Segmentation module, for human body target image and human body target image background to be split;
Target area determining module, for determining target area and obtaining the three-dimensional data of the target area;
Wind area determination module is swept, for determining that air-conditioning sweeps wind region;
Wind angle adjustment module is swept, wind angle is swept for calculating acquisition air-conditioning.
Wherein, image-forming module, including 2 or multiple thermal cameras.
Wherein after air-conditioning unlatching, the image-forming module of infrared binocular camera, it is imaged by detecting infrared source, and By into picture be sent to image preprocessing and correction module, to image carry out noise suppressed, enhancing, utilize the distortion parameter of camera lens Image is carried out the associative operation such as to correct;Then according to the correction parameter of camera calibration, two width that two video cameras are formed Relevant matches position correction in image is on same baseline.Stereo matching is carried out to it.
After matching, by Target Segmentation module, the average background figure for the front and rear multiple image that calculating video camera inputs, Human body target image is split by the difference of human body target image and average background figure.
Target area determining module, according to the human body target image of Target Segmentation module segmentation, target area is determined, wherein The determination of target area includes the determination in single goal region and the determination in multiple target region., will be infrared after determining target area Binocular camera into two target areas marginal point matched and by coordinates matrix calculate obtain target area three Dimension data.
Wind area determination module is swept, the target area quantity determined according to target area determining module and target area Three-dimensional data, it is determined that sweeping the quantity in wind region and that sweeps wind region sweeps wind volume.
Sweep wind angle adjustment module, according to sweep wind area determination module determination sweep wind volume, calculate air-conditioning and sweep wind Angle.
The present embodiment, by image-forming module, image preprocessing and correction module, Target Segmentation module, target area Determining module, wind area determination module is swept, sweep wind angle adjustment module, realized air conditioner intelligent, dynamic and do not need artificial It is default to sweep wind pattern and sweep the function of wind angle.
Example IV
A kind of intelligent air condition stereoscopic vision alignment system is present embodiments provided, as shown in fig. 7, embodiment three is compared, this Embodiment is further improved with lower module:
Further, image preprocessing and correction module include:
Image pre-processing module, for carrying out noise suppressed or enhancing and to human body target image to human body target image Corrected and sent the image corrected to image correction module;
Image correction module, for carrying out human body target image parameter correction according to default uncalibrated image.
Further, target area determining module includes:
Target area determining module, sent for determining target area, and by the target area information determined to three-dimensional Data acquisition module;
Three-dimensional data acquisition module, for determining the three-dimensional data of target area by calculating.
Identification module, for identifying the posture of the human body target;
Selecting module, for the selection pair of the three-dimensional data of the posture according to the human body target and the human body target posture The default air-conditioning answered sweeps wind region.
The identification module, human body target posture is identified, can specifically identify the sitting posture or sleeping position of human body target, For example the sleeping position of human body target is lies low or side is slept,
The wind region area of sweeping of the people to lie low is its area of plane, and the people that side is slept only cannot come from his area of plane Its wind sweeping area is considered, because sweeping wind has only considered that the area of plane can cause its side can not blow to air-conditioner wind, therefore root Wind is swept according to air-conditioning is preset corresponding to the three-dimensional data selection of the human body target under the posture and the posture of different human body targets Region.
Further, sweeping wind area determination module includes:
Quantity determining module, for being determined to sweep the quantity in wind region according to the quantity of target area;
Area determination module, for calculating the external of the target area according to the three-dimensional data of each target area Cuboid, and determine that air-conditioning sweeps wind region according to the circumscribed rectangular body of each target area.
Further, in addition to, merging module, for when the air-conditioning sweep wind region for it is multiple when, will be the multiple Air-conditioning, which is swept wind region merging technique and formed, always sweeps wind region;
Wind velocity module is swept in adjustment, for sweeping wind region according to the air-conditioning always swept in wind region and multiple air-conditionings sweep wind White space between region carries out sweeping wind velocity adjustment.
Wind region is swept by merging air-conditioning, the wind of sweeping for carrying out difference to always sweeping wind region and sweeping wind velocity, the intelligence of raising Energy air-conditioning using and experiencing.
In the present embodiment, by increasing target area determining module, three-dimensional data acquisition module, quantity determining module, area Domain determining module cause infrared binocular stereo vision alignment system it is determined that target area and sweep wind region and sweep wind angle effect Rate is more increased, and determines that result is more accurate so that the intelligent air condition stereoscopic vision alignment system can effectively, intelligence, dynamic The scope of activities of people indoors is monitored, targeting adjustment air-conditioning sweeps wind region and air-conditioning sweeps wind angle.
Specific embodiment described herein is only to spirit explanation for example of the invention.Technology belonging to the present invention is led The technical staff in domain can be made various modifications or supplement to described specific embodiment or be replaced using similar mode Generation, but without departing from the spiritual of the present invention or surmount scope defined in appended claims.

Claims (12)

  1. A kind of 1. method of intelligent air condition stereoscopic vision positioning, it is characterised in that including step:
    S1:Human body target is imaged, and human body target image is pre-processed and corrected;
    S2:According to the human body target image after pretreatment and correction, human body target image and human body target image background are carried out Segmentation;
    S3:According to the human body target image after segmentation, determine target area and obtain the three-dimensional data of the target area;
    S4:According to the three-dimensional data of the target area and target area, determine that air-conditioning sweeps wind region;
    S5:Wind region is swept according to air-conditioning, wind angle is swept by calculating acquisition air-conditioning.
  2. 2. the method for a kind of intelligent air condition stereoscopic vision positioning according to claim 1, it is characterised in that taken the photograph by infrared Camera is imaged to human body target, and the quantity of the thermal camera is for 2 or multiple, wherein, by human body target image and people Body destination image background, which carries out segmentation, includes step:
    Calculate the average background figure of human body target image background;
    Split by the gray scale difference value of human body target image and the average background figure.
  3. 3. the method for a kind of intelligent air condition stereoscopic vision positioning according to claim 2, it is characterised in that obtain the mesh The three-dimensional data in mark region includes step:
    The marginal point of target area is extracted;
    The target area marginal point that different thermal cameras extract is matched and the three-dimensional data of target area is calculated.
  4. A kind of 4. method of intelligent air condition stereoscopic vision positioning according to claim 3, it is characterised in that
    The posture of the human body target is identified, according to the posture of the human body target and the three-dimensional data of the human body target posture Air-conditioning is preset corresponding to selection and sweeps wind region.
  5. 5. the method for a kind of intelligent air condition stereoscopic vision positioning according to claim 4, it is characterised in that determine that air-conditioning is swept Wind region specifically includes step:
    Determined to sweep the quantity in wind region according to the quantity of target area;
    The circumscribed rectangular body of the target area is calculated according to the three-dimensional data of each target area;
    Determine that air-conditioning sweeps wind region according to the circumscribed rectangular body of each target area.
  6. A kind of 6. method of intelligent air condition stereoscopic vision positioning according to claim 5, it is characterised in that
    When the air-conditioning sweep wind region for it is multiple when, the multiple air-conditioning is swept into wind region merging technique into always sweeping wind region;
    Sweep when wind region is always swept described in wind, when sweeping to multiple air-conditionings and sweeping wind region intermediate blank region, adjust this and sweep wind velocity Wind velocity is swept for default blank.
  7. A kind of 7. intelligent air condition stereoscopic vision alignment system, it is characterised in that including:
    Image-forming module:For being imaged to human body target;
    Image preprocessing and correction module, for human body target image to be pre-processed and corrected;
    Target Segmentation module, for human body target image and human body target image background to be split;
    Characteristic area determining module, for determining target area and obtaining the three-dimensional data of the target area;
    Wind area determination module is swept, for determining that air-conditioning sweeps wind region;
    Wind angle adjustment module is swept, wind angle is swept for calculating acquisition air-conditioning.
  8. A kind of 8. intelligent air condition stereoscopic vision alignment system according to claim 7, it is characterised in that image preprocessing and Correction module includes:
    Image pre-processing module, for carrying out noise suppressed or enhancing to human body target image and being carried out to human body target image Correction;
    Image correction module, for carrying out human body target image parameter correction according to default uncalibrated image.
  9. 9. a kind of intelligent air condition stereoscopic vision alignment system according to claim 7, it is characterised in that characteristic area determines Module includes:
    Target area determining module, for determining target area;
    Three-dimensional data acquisition module, for obtaining the three-dimensional data of target area by calculating.
  10. A kind of 10. method of intelligent air condition stereoscopic vision positioning according to claim 9, it is characterised in that including:
    Identification module, for identifying the posture of the human body target;
    Selecting module, for corresponding to the selection of the three-dimensional data of the posture according to the human body target and the human body target posture Default air-conditioning sweeps wind region.
  11. 11. a kind of intelligent air condition stereoscopic vision alignment system according to claim 10, it is characterised in that it is true to sweep wind region Cover half block includes:
    Quantity determining module, for being determined to sweep the quantity in wind region according to the quantity of target area;
    Area determination module, for calculating the circumscribed rectangular of the target area according to the three-dimensional data of each target area Body, and determine that air-conditioning sweeps wind region according to the circumscribed rectangular body of each target area.
  12. 12. a kind of intelligent air condition stereoscopic vision alignment system according to claim 11, it is characterised in that also include:
    Merging module, for when it being multiple that the air-conditioning, which sweeps wind region, the multiple air-conditioning being swept into wind region merging technique and formed always Sweep wind region;
    Wind velocity module is swept in adjustment, for sweeping wind region according to the air-conditioning always swept in wind region and multiple air-conditionings sweep wind region Between white space sweep wind velocity adjustment.
CN201710656194.4A 2017-08-03 2017-08-03 A kind of method and system of intelligent air condition stereoscopic vision positioning Pending CN107514745A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710656194.4A CN107514745A (en) 2017-08-03 2017-08-03 A kind of method and system of intelligent air condition stereoscopic vision positioning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710656194.4A CN107514745A (en) 2017-08-03 2017-08-03 A kind of method and system of intelligent air condition stereoscopic vision positioning

Publications (1)

Publication Number Publication Date
CN107514745A true CN107514745A (en) 2017-12-26

Family

ID=60721969

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710656194.4A Pending CN107514745A (en) 2017-08-03 2017-08-03 A kind of method and system of intelligent air condition stereoscopic vision positioning

Country Status (1)

Country Link
CN (1) CN107514745A (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109725609A (en) * 2018-11-12 2019-05-07 于怡然 A kind of intelligent air-conditioning system
CN110068120A (en) * 2019-04-23 2019-07-30 广东美博制冷设备有限公司 A kind of refrigeration control method and device of air conditioner
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 and air-conditioning, storage medium
CN112178865A (en) * 2020-08-17 2021-01-05 珠海格力电器股份有限公司 Air conditioner pollutant detection method, purification method, control method and air conditioner
CN112178864A (en) * 2020-08-17 2021-01-05 珠海格力电器股份有限公司 Air conditioner pollutant detection method, control method and air conditioner
CN112460756A (en) * 2020-11-09 2021-03-09 青岛海尔空调电子有限公司 Method and device for determining air outlet area of indoor unit of air conditioner

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103090503A (en) * 2011-10-27 2013-05-08 海尔集团公司 Air-conditioning device and control method thereof
CN103714345A (en) * 2013-12-27 2014-04-09 Tcl集团股份有限公司 Method and system for detecting fingertip space position based on binocular stereoscopic vision
CN103968507A (en) * 2014-05-20 2014-08-06 珠海格力电器股份有限公司 Air conditioner and air conditioner control method and system
CN104697112A (en) * 2015-03-04 2015-06-10 广东美的制冷设备有限公司 Control method and system of air conditioner
CN105020867A (en) * 2015-08-10 2015-11-04 珠海格力电器股份有限公司 Air conditioner control method, air conditioner and human body detection device
CN105258293A (en) * 2015-10-29 2016-01-20 广东美的制冷设备有限公司 Air output control method of air conditioner
CN106503605A (en) * 2015-09-01 2017-03-15 南京理工大学 Human body target recognition methods based on stereovision technique
CN106991410A (en) * 2017-04-17 2017-07-28 合肥酷庆信息科技有限公司 A kind of railway foreign body intrusion distributed intelligent monitoring system based on machine vision

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103090503A (en) * 2011-10-27 2013-05-08 海尔集团公司 Air-conditioning device and control method thereof
CN103714345A (en) * 2013-12-27 2014-04-09 Tcl集团股份有限公司 Method and system for detecting fingertip space position based on binocular stereoscopic vision
CN103968507A (en) * 2014-05-20 2014-08-06 珠海格力电器股份有限公司 Air conditioner and air conditioner control method and system
CN104697112A (en) * 2015-03-04 2015-06-10 广东美的制冷设备有限公司 Control method and system of air conditioner
CN105020867A (en) * 2015-08-10 2015-11-04 珠海格力电器股份有限公司 Air conditioner control method, air conditioner and human body detection device
CN106503605A (en) * 2015-09-01 2017-03-15 南京理工大学 Human body target recognition methods based on stereovision technique
CN105258293A (en) * 2015-10-29 2016-01-20 广东美的制冷设备有限公司 Air output control method of air conditioner
CN106991410A (en) * 2017-04-17 2017-07-28 合肥酷庆信息科技有限公司 A kind of railway foreign body intrusion distributed intelligent monitoring system based on machine vision

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109725609A (en) * 2018-11-12 2019-05-07 于怡然 A kind of intelligent air-conditioning system
CN109725609B (en) * 2018-11-12 2022-08-09 广州浩科捷睿科技有限责任公司 Intelligent air conditioning system
CN110068120A (en) * 2019-04-23 2019-07-30 广东美博制冷设备有限公司 A kind of refrigeration control method and device of air conditioner
CN110068120B (en) * 2019-04-23 2021-08-27 广东美博制冷设备有限公司 Refrigeration control method and device of air conditioner
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 and air-conditioning, storage medium
CN112178865A (en) * 2020-08-17 2021-01-05 珠海格力电器股份有限公司 Air conditioner pollutant detection method, purification method, control method and air conditioner
CN112178864A (en) * 2020-08-17 2021-01-05 珠海格力电器股份有限公司 Air conditioner pollutant detection method, control method and air conditioner
CN112178864B (en) * 2020-08-17 2021-11-23 珠海格力电器股份有限公司 Air conditioner pollutant detection method, control method and air conditioner
CN112178865B (en) * 2020-08-17 2022-04-01 珠海格力电器股份有限公司 Air conditioner pollutant detection method, purification method, control method and air conditioner
CN112460756A (en) * 2020-11-09 2021-03-09 青岛海尔空调电子有限公司 Method and device for determining air outlet area of indoor unit of air conditioner

Similar Documents

Publication Publication Date Title
CN107514745A (en) A kind of method and system of intelligent air condition stereoscopic vision positioning
WO2022142759A1 (en) Lidar and camera joint calibration method
CN108765498A (en) Monocular vision tracking, device and storage medium
JP6295645B2 (en) Object detection method and object detection apparatus
CN113436260B (en) Mobile robot pose estimation method and system based on multi-sensor tight coupling
CN102830793B (en) Sight tracing and equipment
WO2017080108A1 (en) Flying device, flying control system and method
CN106772431A (en) A kind of Depth Information Acquistion devices and methods therefor of combination TOF technologies and binocular vision
CN110232389A (en) A kind of stereoscopic vision air navigation aid based on green crop feature extraction invariance
CN109360240A (en) A kind of small drone localization method based on binocular vision
CN105574905B (en) A kind of two dimensional image expression method of three-dimensional laser point cloud data
CN111612845A (en) Laser radar and camera combined calibration method based on mobile calibration plate
CN103886107B (en) Robot localization and map structuring system based on ceiling image information
TW202004560A (en) Object detection system, autonomous vehicle, and object detection method thereof
CN107239748A (en) Robot target identification and localization method based on gridiron pattern calibration technique
CN109145864A (en) Determine method, apparatus, storage medium and the terminal device of visibility region
CN108731587A (en) A kind of the unmanned plane dynamic target tracking and localization method of view-based access control model
CN102930252A (en) Sight tracking method based on neural network head movement compensation
CN111429533B (en) Camera lens distortion parameter estimation device and method
CN110657785A (en) Efficient scene depth information acquisition method and system
CN110136202A (en) A kind of multi-targets recognition and localization method based on SSD and dual camera
CN105631859B (en) Three-degree-of-freedom bionic stereo visual system
CN102831601A (en) Three-dimensional matching method based on union similarity measure and self-adaptive support weighting
CN106537217B (en) Wide pentrution imaging
CN106650701A (en) Binocular vision-based method and apparatus for detecting barrier in indoor shadow environment

Legal Events

Date Code Title Description
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: 20171226