CN105180385A - Method and system for controlling air conditioner based on facial features and air conditioner - Google Patents
Method and system for controlling air conditioner based on facial features and air conditioner Download PDFInfo
- Publication number
- CN105180385A CN105180385A CN201510733881.2A CN201510733881A CN105180385A CN 105180385 A CN105180385 A CN 105180385A CN 201510733881 A CN201510733881 A CN 201510733881A CN 105180385 A CN105180385 A CN 105180385A
- Authority
- CN
- China
- Prior art keywords
- face
- degree
- protected
- picture
- heat
- 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.)
- Withdrawn
Links
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
- F24F11/30—Control or safety arrangements for purposes related to the operation of the system, e.g. for safety or monitoring
-
- 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
-
- 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
-
- 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
- F24F11/64—Electronic processing using pre-stored data
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Signal Processing (AREA)
- Air Conditioning Control Device (AREA)
Abstract
The invention relates to a method and system for controlling an air conditioner based on facial features. The method comprises receiving an image containing a face, and analyzing facial features according to the image to obtain parameters of the facial features; calculating the fat degree RF of the face according to the parameters of the facial features; calculating heat resistant degree L according to the fat degree RF of the face; and generating a control instruction used for modulating operation parameters of an air conditioner according to the heat resistant degree L. The method can analyzes air conditioner temperature preference of a user in real time using a common face image. Compared with the prior art, a pre-established face database for recognizing the user identity is needless, determination about activity can be made without analysis on a plurality of images, a special temperature sensing camera is needless, a common camera of a common smart phone can be used for detecting temperature preference. The method and system for controlling an air conditioner based on facial features are superior to the prior art in develop cost of goods, user privacy, computational complexity of hardware, and places of applications and the like.
Description
Technical field
The present invention relates to airconditioning control field, particularly relate to and a kind ofly carry out the method for airconditioning control, system and air-conditioning based on face feature.
Background technology
Existed in the market some can intelligent control air-conditioning run method.Some method is mated by the image its memory module of the face image of extraction and air-conditioning stored, according to matching result adjustment air conditioner operation parameters, although this method can control operation of air conditioner by face recognition, but need the corresponding relation set up out in advance between face database and air-conditioning parameter, to obtain the air-conditioning preference of user, this method will do more preparation early stage, is unfavorable for realizing; Also have some by detecting face location and size, and then calculate the amount of movement of indoor people, calculated the activity of people, carry out airconditioning control according to activity by amount of movement, this method needs a large amount of picture continuously of acquisition to analyze, and amount of calculation is very large; Also have some by infrared camera acquisitor surface temperature image or picture, control operation of air conditioner according to human surface temperature's image or picture, this method need increase the cost of product.
Summary of the invention
Technical problem to be solved by this invention is for the deficiencies in the prior art, provides a kind of and carries out the method for airconditioning control, system and air-conditioning based on face feature.
The technical scheme that the present invention solves the problems of the technologies described above is as follows:
Carry out a method for airconditioning control based on face feature, comprise the steps:
Step 1, receives the picture that has face, carries out face feature analysis according to described picture, obtains face feature parameter;
Step 2, calculates the fat or thin degree RF of face according to described face feature parameter;
Step 3, according to the fat or thin degree RF of the described face degree L that calculates that To Be Protected from Heat;
Step 4, produces the control instruction of adjustment air conditioner operation parameters according to described To Be Protected from Heat degree L.
The invention has the beneficial effects as follows: the present invention utilizes a common face picture, and instant analysis goes out the air-conditioner temperature preference of user.Compared with prior art, without the need to setting up face database identification user identity in advance, not needing to analyze plurality of pictures and could produce activity judgement, being particularly suitable for public space, getting together temporarily, be inconvenient to the occasion collecting face picture library; The present invention does not need to use distinct temperature sensing pickup camera yet, uses the common lens of general intelligence mobile phone can complete the detecting of cold and hot preference.All obviously prior art is better than in commodity development cost, privacy of user, hardware amount of calculation and application scenarios etc.
On the basis of technique scheme, the present invention can also do following improvement.
Further, being implemented as of step 2:
Rapid 2.1, select at least one facial feature points, calculated by the facial feature points selected and cut reference point, the above picture excision of reference point will be cut, retain residue face picture;
Step 2.2, analyzes the face's picture after cutting, and obtains fat or thin degree RF, the RF=skin pixels quantity/picture pixels quantity of face by the ratio calculating skin pixels quantity and picture pixels quantity.
Adopting the beneficial effect of above-mentioned further scheme to be: by cutting picture, effectively to avoid top shape of face to affect the fat or thin degree analyzing of face, analyzing cheek and chin skin pixels area that fat is the most easily accumulated by face, increasing fat or thin degree identification accuracy.
Further, the facial feature points of selection calculates with weighting or average mode and cuts reference point by described step 2.1; The facial feature points of described selection is the characteristic point of ocular vicinity.
Adopt the beneficial effect of above-mentioned further scheme: hair style may affect the judgement of the fat or thin degree of face, the hair style of people can not hide to eyes by analysis, therefore the characteristic point (Landmarks) selecting ocular vicinity is benchmark, picture is cut, obtain the picture that face cuts below reference point, with the analysis effectively avoiding hair style to affect the fat or thin degree of face; And body fat easily accumulates on face both sides and chin, therefore acquisition cuts the picture of below reference point, effectively can analyze face's fat degree.
Further, being implemented as of step 3:
Step 3.1, converts fat or thin for face degree RF to expection body fat rate EBFR by the mode of data normalization;
Step 3.2, calculates rest energy expenditure RDEE every day according to expection body fat rate EBFR and average batheroom scale;
Step 3.3, by rest energy expenditure RDEE every day is normalized between 0 to 1, the degree L that obtains that To Be Protected from Heat.
The beneficial effect of above-mentioned further scheme is adopted to be: fat or thin for face degree RF to be converted to expection body fat rate EBFR, rest energy expenditure RDEE every day is calculated according to expection body fat rate EBFR and average batheroom scale, RDEE is normalized to the numerical value between 0 and 1, represent a people To Be Protected from Heat degree, represent close to 0 by the moon and be more afraid of cold, more represent that more To Be Protected from Heat close to 1, the degree of being afraid of cold that To Be Protected from Heat of a people can be reacted accurately and intuitively.
Further, expect described in step 3.1 that the computing formula of body fat rate EBFR is as follows:
EBFR＝(Rb-Ra)x[(RF-N)/(M-N)]+Ra；
Wherein, described Rb is the body fat rate upper limit, and Ra is body fat rate lower limit, and RF represents the fat or thin degree of face, and N represents the fat or thin degree lower limit of face, and M represents the fat or thin degree upper limit of face;
Described Rb, Ra are according to the presence or absence of gender data, and the gender data when determining sex is determined, have for:
Sex (women s=0, male sex s=1, countless according to s=2);
IfRF<NorRF>M, cannot from picture identification BFM;
If (s==0) Ra=women body fat rate lower limit, Rb=women's body fat rate higher limit;
ElseIf (s==1) Ra=male sex body fat rate lower limit, Rb=male sex's body fat rate higher limit;
ElseRa=gives tacit consent to body fat rate lower limit, and Rb=gives tacit consent to body fat rate higher limit;
Every day described in step 3.2, the computing formula of rest energy expenditure RDEE was as follows:
RDEE＝370+21.6x(1－EBFR)xKg；
Wherein, Kg is average weight, when differentiating age level and/or sex by picture, then calculates according to compatriots' average weight, if go out age level and/or sex by picture identifiable design, then calculates average weight according to this age level;
The computing formula of described in step 3.3, To Be Protected from Heat degree L is as follows:
IF(RDEE<Ea)L＝0；
ELSEIF(RDEE>Eb)L＝1；
ELSEL＝(RDEE-Ea)/(Eb-Ea)；
Wherein, Ea is the lower bound of rest energy expenditure RDEE every day, and Eb is the upper bound of rest energy expenditure RDEE every day;
Ea＝370+21.6x(1-Rb)xKa；Eb370+21.6x(1-Ra)xKb；
Wherein, Ka is body weight lower bound, and Kb is the body weight upper bound.
Further, the specific implementation of step 4 comprises:
According to described To Be Protected from Heat degree adjustment air-conditioner temperature, humidity, air quantity, Auto On Time point/Auto Power On distance, automatic shutdown time point/automatic shutdown distance, continuous temperature change curve and User Interface one or more; Wherein said Auto Power On Distance geometry automatic shutdown distance all refers to the distance between user to air-conditioning equipment, when distance between user and air-conditioning is less than or equal to Auto Power On distance, Auto Power On, when the distance between user and air-conditioning is greater than automatic shutdown distance, automatic shutdown;
Also comprise and carry out colony's air-conditioning parameter optimization adjustment, be specially:
When the pictures received has multiple face, calculate each face respectively To Be Protected from Heat degree, by everyone, To Be Protected from Heat that degree is weighted or average computation, draws colony To Be Protected from Heat degree, in conjunction with colony's To Be Protected from Heat degree adjustment air conditioner operation parameters.
Adopt the beneficial effect of above-mentioned further scheme to be: degree can adjust the multiple operational factor of air-conditioning according to To Be Protected from Heat, and gives the method being suitable for colony's airconditioning control, this method is very applicable in occasions such as public space, interim parties.
Carry out a system for airconditioning control based on face feature, comprise face feature analysis module, the fat or thin degree computing module of face, To Be Protected from Heat degree computing module and air conditioner operation parameters adjusting module; Described face feature analysis module, for receiving the picture that has face, carries out face feature analysis according to described picture, obtains face feature parameter; The fat or thin degree computing module of described face, for calculating the fat or thin degree RF of face according to described face feature parameter; Described To Be Protected from Heat degree computing module, for the degree L that calculates according to the fat or thin degree RF of described face that To Be Protected from Heat; Described air conditioner operation parameters adjusting module, for producing the control instruction of adjustment air conditioner operation parameters according to described To Be Protected from Heat degree L.
On the basis of technique scheme, the present invention can also do following improvement.
Further, described face fat or thin degree computing module comprises and cuts the fat or thin degree computing unit of unit and face; Describedly cutting unit, for cutting described picture according to face feature parameter, obtaining face's picture of below eyes; The fat or thin degree computing unit of described face, for analyzing the picture after cutting, obtains fat or thin degree RF, the RF=skin pixels quantity/picture pixels quantity of face by the ratio calculating skin pixels quantity and picture pixels quantity.
Further, described To Be Protected from Heat degree computing module comprises expection body fat rate computing unit, every day rest energy expenditure computing unit and To Be Protected from Heat degree computing unit; Described expection body fat rate computing unit, converts fat or thin for face degree RF to expection body fat rate EBFR for the mode by data normalization; Described every day rest energy expenditure computing unit, for calculating rest energy expenditure RDEE every day according to expection body fat rate EBFR and average batheroom scale; Described To Be Protected from Heat degree computing unit, by rest energy expenditure RDEE every day is normalized between 0 to 1, the degree L that obtains that To Be Protected from Heat.
A kind of air-conditioning, comprises the system of carrying out airconditioning control based on face feature described in technique scheme.
Accompanying drawing explanation
Fig. 1 is a kind of method flow diagram carrying out airconditioning control based on face feature of the present invention;
Fig. 2 a is schematic diagram before cropping of the present invention;
Fig. 2 b is schematic diagram after cropping of the present invention;
Fig. 3 is a kind of system block diagram carrying out airconditioning control based on face feature of the present invention.
Detailed description of the invention
Be described principle of the present invention and feature below in conjunction with accompanying drawing, example, only for explaining the present invention, is not intended to limit scope of the present invention.
As shown in Figure 1, a kind of method of carrying out airconditioning control based on face feature, comprises the steps:
(1) face feature analysis:
Step 1, receives the picture that has face, carries out face feature analysis according to described picture, obtains face feature parameter.Particularly, described face feature parameter comprises face's scope in picture, eyebrow position, eye position, nose position and face position.
(2) the fat or thin degree of face calculates:
Due to a people get fat time, face not easily accumulate fat, and size can not change, but face's Lower Half profile easily accumulates fat to chin.Therefore can pass through the ratio that whole face shared by analyzing skin area is long-pending, calculate the fat or thin degree of face.
Step 2, calculates the fat or thin degree RF of face according to described face feature parameter.
Particularly, being implemented as of step 2:
Rapid 2.1, select at least one facial feature points, calculated by the facial feature points selected and cut reference point, the above picture excision of reference point will be cut, retain residue face picture; Described facial feature points comprises:
A. outside left eyebrow, inside left eyebrow, outside left eye upper limb, left eye, inside left eye, above left eye lower edge, pupil of left eye, bridge of the nose left border, the left wing of nose, outside the left wing of nose;
B. outside right eyebrow, inside right eyebrow, outside right eye upper limb, right eye, inside right eye, above right eye lower edge, pupil of right eye, bridge of the nose right border, the right wing of nose, outside the right wing of nose;
C. nose;
D. upper lip upper limb, upper lip lower edge, lower lip upper limb, lower lip lower edge, face left border, face right border;
By above characteristic point, to combine or weighting produces cuts reference point, all can obtain the picture that face Lower Half retains chin portion after cutting;
By cutting picture, effectively avoiding top shape of face to affect the fat or thin degree analyzing of face, analyzing cheek and chin skin pixels area that fat is the most easily accumulated by face, increasing fat or thin degree identification accuracy.
The facial feature points of selection calculates with weighting or average mode and cuts reference point by described step 2.1; The facial feature points of described selection is the characteristic point of ocular vicinity.
Hair style may affect the judgement of the fat or thin degree of face, the hair style of people can not hide to eyes by analysis, therefore the characteristic point (Landmarks) selecting ocular vicinity is benchmark, picture is cut, obtain the picture that face cuts below reference point, with the analysis effectively avoiding hair style to affect the fat or thin degree of face; And body fat easily accumulates on face both sides and chin, therefore acquisition cuts the picture of below reference point, effectively can analyze face's fat degree.
Illustrate below in conjunction with instantiation:
As shown in Fig. 2 a, 2b, pupil of left eye and pupil of right eye two characteristic points are selected to be used as reference point.The position of pupil of left eye position (x, y) is (100,210), and pupil of right eye position is (220,210), and the average height position of getting two is used as and is cut reference point: (210+210)/2=210.Therefore, cropping is become to retain the picture that coordinate y is greater than 210.
Step 2.2, analyzes the face's picture after cutting, and obtains fat or thin degree RF, the RF=skin pixels quantity/picture pixels quantity of face by the ratio calculating skin pixels quantity and picture pixels quantity.
Picture after cutting is analyzed, analyzes face's obese degree.Because face chin adipose tissue makes cutis laxa, therefore can pass through and measure the ratio that skin accounts for whole pictures, carry out the estimation of fat or thin degree.
Whether may be skin, existing many academic research achievements if detecting a pixel.Generally acknowledge that method is that pixel is gone to HSV colour gamut from RGB colour gamut (ColorSpace) more accurately, and the H value analyzing pixel is whether in particular range.Many achievements in research count, when H value is 6 ~ 38, closest to the color of skin.Therefore, after scanning picture, can obtain pixel is the quantity of skin color and the total quantity of picture pixels.For the people of obesity, because chin surrounding skin is more lax, therefore under same photo environment, higher skin pixels quantity can be obtained.
Through above principle, one can be designed and weigh the fat or thin index of face, the fat or thin degree (FaceFatRatio, RF) of face:
RF=skin pixels quantity (skinpixels)/picture pixels quantity (pixels).
(3) degree that To Be Protected from Heat calculates:
Step 3, according to the fat or thin degree RF of the described face degree L that calculates that To Be Protected from Heat.
The health of people is a constant temperature system, and body temperature is the heat that metabolism produces.Somagenic need keeps constant temperature, and therefore when metabolism amount is larger, hotness can be sent to hypothalamus by the temperature induction mechanism of skin, and the hypothalamus being responsible for controlling constant temperature can send sweats, and angiectatic control command, to help heat dissipating body.Therefore metabolism people faster, as man, child and sportsman easily feel heat, the group that belongs to that To Be Protected from Heat.The people that woman, old man and muscle ratios are less, because metabolism amount is less, metabolic product heat energy is lower, therefore in order to maintain health constant temperature, easily experiencing cold, belonging to the group be comparatively afraid of cold.Therefore the amount weighing body intracellular metabolic is that To Be Protected from Heat or the important evidence of degree of being afraid of cold in estimation.
Measure the method for human metabolism, studied and exceeded a century, more famous prediction mode is 1919, the people such as scholar Harris-Benedict, the formula proposed.Utilize height, body weight, sex, age, estimate basic metabolism every day (Basalmetabolicrate, BMR) of a people.Have passed through century-old research and development, comparatively accept extensively at present, use body fat rate to infer the basic metabolism of people.Rest energy expenditure (RestingDailyEnergyExpenditure, RDEE) algorithm every day that Katch-McArdle proposes is as follows:
RDEE=370+21.6x (100%-body fat rate) x body weight (kg);
Therefore, can pass through body fat rate and body weight, calculate every D aily energy expenditure, namely people produces the degree of heat energy.In June, 2015, People's Republic of China's national health and Family Planning Committee publish " Chinese residents nourishment and chronic disease status report (2015) ", report content is pointed out, the average weight of 18 years old, the whole nation and above adult male and women is respectively 66.2kg and 57.3kg.
The sex ratio that State Statistics Bureau issued in 2015, male demographic 70,079 ten thousand people, female population 66,703 ten thousand people, male demographic 3,376 ten thousand people more than women, total population sex ratio is 105.06 (with women for 100).By men and women's average weight through after M-F weighted calculation, compatriots' average weight: (105.06x66.2+100x57.3)/(105.06+100)=61.86 kilogram.Therefore can utilize the method for statistics, obtain the population mean body weight of masculinity and femininity.
According to experimental result of the present invention, the distribution of the fat or thin degree of face (RF), between [0.2-0.98], is generally picture modified lower than 0.2 many, without the colour of skin.Higher than 0.98 be background noise impact.And the body fat of common people is approximately from 6 to 40.Therefore can pass through the mode of following data normalization, by fat or thin for face ratio, convert expection body fat rate to.Expection body fat rate (ExpectedBodyFatRatio, EBFR).After obtaining EBFR and average weight, can estimate to calculate possible rest energy expenditure RDEE every day of a people, account form is as follows:
Input: fat or thin degree (RF), sex (women s=0, male sex s=1, countless according to s=2);
Output: rest energy expenditure every day (Lh);
Ra: body fat rate lower bound, Rb: the body fat rate upper bound, Kg: average weight;
IfRF<0.2orRF>0.98, cannot from picture identification BFM;
If(s＝＝0)Ra＝10％,Rb＝45％；
ElseIf(s＝＝1)Ra＝6％,Rb＝35％；
ElseRa＝8％,Rb＝40％；
EBFR＝(Rb-Ra)x[(RF-0.2)/(0.98-0.2)]+Ra；
RDEE＝370+21.6x(1－EBFR)xKg。
During RDEE calculates, body weight Kg is statistics, can pass through national statistics report and obtains, if when image identifiable design goes out age level or sex, can the body weight further corresponding to this age level calculate more accurately.In the present embodiment, N value is 0.2, M value is 0.98.
Change through above data normalization, the rest energy expenditure every day (RestingDailyEnergyExpenditure, RDEE) of a facial photo can be drawn.The not accurate every D aily energy expenditure of these data, but be enough to expression human metabolism and the relative extent producing body heat.
The body fat rate of common people and the distribution of body weight are haply: body fat rate lower bound is Ra%, body fat rate upper bound Rb%, body weight lower bound Ka, body weight upper bound Kb, after bringing RDEE formula into, can show that RDEE scope [Ea, Eb] is: [370+21.6x (1-Rb) xKa, 370+21.6x (1-Ra) xKb].
If when such as implementing, by Ra=8%, Rb=40%, Ka=40, Kb=120, then can calculate [Ea, Eb]=[888,2755].
About setting body weight lower bound Ka and body weight upper bound Kb: because body weight is not normal distribution, if do not set the upper bound and the lower bound of this value from the angle of statistics, statistical special abnormality value (Outlier) sample will be made to affect result of calculation.Such as the body weight upper bound is set as 300 jin, the body weight of common people can be made to distribute and to seem Relatively centralized, and then make the RDEE difference of common people minimum, it is minimum that the air-conditioner temperature that common people are calculated controls difference.Large to make final air-conditioner temperature control difference, larger body weight lower bound can be set, or the less body weight upper bound, make the RDEE difference of individual user large, air-conditioner temperature changes larger.。
Because RDEE is the degree that human body produces heat, health holds constant temperature, the people that therefore RDEE is higher because of needs, more easily feels hot, makes health start the cooling mechanism such as perspire and blood vessel dilatation.Therefore can pass through the height of RDEE, weigh a people and be afraid of cold and the degree that To Be Protected from Heat.By RDEE standardization (Normalization) to the numerical value between 0 and 1, represent a people To Be Protected from Heat degree, more represent close to 0 and be more afraid of cold, more represent that more To Be Protected from Heat close to 1.
Be afraid of that Hot degree L is calculated as follows:
Input: rest energy expenditure RDEE every day of user;
Output: be afraid of Hot degree (L);
Controllable parameter: RDEE lower bound (Ea), the RDEE upper bound (Eb);
IF(RDEE<Ea)L＝0；
ELSEIF(RDEE>Eb)L＝1；
ELSEL＝(RDEE-Ea)/(Eb-Ea)。
Be afraid of that Hot degree (L) has reacted the health heat production degree of a people, the higher health of L more easily feels hot, and L is lower, and health more easily feels a sense of cold.
(4) operation of air conditioner is adjusted
Step 4, produces the control instruction of adjustment air conditioner operation parameters according to described To Be Protected from Heat degree L.
Particularly, according to described To Be Protected from Heat degree adjustment air-conditioner temperature, humidity, air quantity, Auto On Time point/Auto Power On distance, automatic shutdown time point/automatic shutdown distance, continuous temperature change curve and User Interface one or more; Wherein said Auto Power On Distance geometry automatic shutdown distance all refers to the distance between user to air-conditioning equipment, when distance between user and air-conditioning is less than or equal to Auto Power On distance, Auto Power On, when the distance between user and air-conditioning is greater than automatic shutdown distance, automatic shutdown.
Also comprise and carry out colony's air-conditioning parameter optimization adjustment, be specially:
When the pictures received has multiple face, calculate each face respectively To Be Protected from Heat degree, by everyone, To Be Protected from Heat that degree is weighted or average computation, draws colony To Be Protected from Heat degree, in conjunction with colony's To Be Protected from Heat degree adjustment air conditioner operation parameters.
A. colony's air-conditioner temperature optimization:
When friend gets together, clap a group photograph, utilize mobile phone A PP, upload group photograph to air-conditioning server, after server calculates everyone To Be Protected from Heat degree, by everyone, To Be Protected from Heat that degree is weighted, or average computation, draw colony To Be Protected from Heat and give degree this colony's optimal comfort temperature, meet colony's maximal comfort.
B. the start being afraid of Hot degree is combined:
Timing start-up, Auto Power On or Auto On Mod of going home, can be afraid of Hot degree by root According, the time point of adjustment start.To Be Protected from Heat that degree is higher for user, can start shooting ahead of time, make the environment that the user that To Be Protected from Heat can comparatively be felt nice and cool.
Auto On Time=Auto On Time-To Be Protected from Heat degree xF.Wherein F is systematic parameter, or a variable function, is used for adjusting the time point of starting shooting ahead of time because To Be Protected from Heat.Therefore, the people that more To Be Protected from Heat can obtain the air-conditioning available machine time comparatively early, reaches preferably refrigerating capacity.
Auto On Mod of going home to have the point or threshold, such as distance or smart machine wireless signal strength on an opportunity driving air-conditioning start usually.Can according to To Be Protected from Heat degree correction start threshold.If set close to the start of 20 meters, room, then according to To Be Protected from Heat degree, can increase or reduce this distance threshold.Such as: start distance threshold=20+20x To Be Protected from Heat degree (L).Therefore, the people that more To Be Protected from Heat can obtain good cooling feeling, the people that more To Be Protected from Heat, and start is comparatively slow, can get over power saving.
C. the shutdown of To Be Protected from Heat degree is combined:
Automatic shutdown time=automatic shutdown time+To Be Protected from Heat degree xF.Wherein F is systematic parameter or a variable function, is used for adjusting the time point extending shutdown because To Be Protected from Heat.Therefore, the people that more To Be Protected from Heat can obtain the more late air-conditioning unused time.Such as sleep timing shutdown pattern; often shut down after setting n hour; the automatic shutdown way example of the degree that make use of that To Be Protected from Heat is: turning device to sleep mode time=user sets hourage+Cx To Be Protected from Heat degree (L); if C is set as 0.5; degree that To Be Protected from Heat is 1, then sleeping, automatic shutdown time can To Be Protected from Heat and delay 0.5 hour because of user.
D. the temperature adjustment of To Be Protected from Heat degree is combined:
Intelligent air condition has continuous temperature control model often, and sleep curve model controls the variations in temperature of 8 ~ 10 hours lengths of one's sleep, healthy heating mode, gradually by temperature to outdoor temperature.The change of this kind of continuous temperature controls, degree of can applying mechanically that To Be Protected from Heat.When To Be Protected from Heat, degree is high, heats up slower, lowers the temperature faster.Make air-conditioning can do exquisiter temperature according to physique to control.
For the sleep curve of continuous eight hours: first three hour, cooling twice, latter two hour, heats up three degree.The degree that can utilize that To Be Protected from Heat, is optimized amendment to upper curve: lower the temperature 2 degree time=3 – Cx To Be Protected from Heat degree (L), and heat up 3 degree time=2+Dx To Be Protected from Heat degree (L), wherein C and D is a regulatable function.The people that degree of so can reaching that To Be Protected from Heat is higher, lower the temperature faster, heat up slower effect.This mode can be applicable to arbitrary continuation temperature and regulates, and continuous temperature is regulated can To Be Protected from Heat that degree is optimized according to people.
E. user interactions interface is optimized:
The people that To Be Protected from Heat, needs to lower the temperature faster, can according to this demand dynamic conditioning user interactions interface.When using smart machine APP operating air conditioner, the people that more To Be Protected from Heat, refrigeration control interface dynamic tunes up, and allows the people Vietnamese side that more To Be Protected from Heat just freeze.In addition also can adjust from the degree of cooling.General air-conditioning temperature-reducing is 0.5 degree to 1 degree.Can according to the degree that To Be Protected from Heat, adjustment cooling meter full scale.
Below will get together for friend temporarily, illustrate that the operation of air conditioner how completed based on face feature by the present invention is controlled.
The user A male sex, has invited user B women, has dinner to family.Take two people through mobile phone to take a group photo souvenir.Utilize mobile phone that two people's group photos are sent to Internet of Things air-conditioning server afterwards.The facial photo of user A and user B was not all sent to server, and the photo that server does not deposit any user is used for identifying identity also there is not the air-conditioning use habit of user B.The customized air themperature of this kind of interim guest, patent in the past and technology cannot be reached.The implementation step that the present invention reaches the customized air themperature of interim guest is as follows:
1. face feature analysis
Server is through image processing mode, and analyzing this photo has two faces, produces the square type position of face in picture, pupil of both eyes position.Existing open source software or cloud service can complete this analysis at present.Such as OpenCV can complete face location detecting and pupil position detecting.Microsoft's cloud service provides face's gender analysis and Analysis of age.Therefore, can to produce the analysis result of this photo as follows for server:
Face 1: the male sex, 28 years old, face's square type scope 1, pupil of both eyes position 1.
Face 2: women, 24 years old, face's square type scope 1, pupil of both eyes position 2.
2. the fat or thin degree of face calculates
Cut face 1 and face 2 figure phase, remain in pupil position with lower part, obtain generation two pictures p1 and p2.P1 and p1 is converted to HSV color gamut space by RGB color gamut space.The each pixel of scanning p1 and p2, if the h value of pixel is between 6-38, is represented as skin.The fat or thin degree (RatioofFat, RF) of Liang Zhang face=skin pixels quantity (skinpixels)/picture pixels quantity (pixels).Two values can be obtained.Suppose that the RF calculated through p1, p2 is as follows:
RF1＝0.72；
RF2＝0.48。
3. degree that To Be Protected from Heat calculates
Expection body fat rate:
The average weight of masculinity and femininity is respectively 66.2kg and 57.3kg, women Ra=10%, Rb=45, male sex Ra=6%, Rb=35, therefore:
EBFR1＝(Rb-Ra)x[(RF1-0.2)/(0.98-0.2)]+Ra＝[(35-6)x(0.72-0.2)/0.78]+6＝25.33％＝0.2533；
EBFR2＝(Rb-Ra)x[(RF2-0.2)/(0.98-0.2)]+Ra＝[(45-10)x(0.48-0.2)/0.78]+10＝22.56％＝0.2256；
Every day rest energy expenditure (RestingDailyEnergyExpenditure, RDEE) RDEE=370+21.6x (1-EBFR) xKg;
RDEE1＝370+21.6x(1－0.2533)x66.2＝1437.7；
RDEE2＝370+21.6x(1－0.2256)x57.3＝1328.5；
To Be Protected from Heat the degree (L) of calculating:
Parameter when first initialization system is implemented: Ra=6%, Rb=40%, Ka=40, Kb=120, then can calculate [Ea, Eb]=[888,2806].
L1＝(RDEE1-Ea)/(Eb-Ea)＝(1437.7-888)/(2806-888)＝0.287；
L2＝(RDEE2-Ea)/(Eb-Ea)＝(1328.5-888)/(2806-888)＝0.230；
Because To Be Protected from Heat degree is more close to 0, represents and be more afraid of cold, in this example, calculate this two people and all belong to the state of being comparatively afraid of cold.
4. adjust operation of air conditioner
Because of in photo for this reason, detect the face more than a people, therefore, start colony's degree of being afraid of cold and calculate.Through weighting or average mode, the arrangement calculated under colony many people environment is afraid of cold degree, and this example uses average mode to calculate colony and to be afraid of cold degree.Group's degree of being afraid of cold calculates, Lg=L1+L2/2=(0.287+0.230)/2=0.259.
To be afraid of cold degree Lg according to colony, an embodiment of adjustment air-conditioner temperature, select air-conditioning adjustable strategies: " more To Be Protected from Heat; more close to 24 degree; be more afraid of cold, more close to 30 degree ", temperature exports To=30 – [(30-24) * Lg], through above strategy, the most applicable temperature that can calculate this two people is: 30-(6*0.259)=28.446 degree.Suppose that the temperature Adjustment precision of this air-conditioning is 1 degree.Therefore through the mode rounded up, this temperature transition is become 28 degree.Finally, air-conditioning exports 28 degree of this two people the most applicable.
As shown in Figure 3, a kind of system of carrying out airconditioning control based on face feature, comprises face feature analysis module, the fat or thin degree computing module of face, To Be Protected from Heat degree computing module and air conditioner operation parameters adjusting module; Described face feature analysis module, for receiving the picture that has face, carries out face feature analysis according to described picture, obtains face feature parameter; The fat or thin degree computing module of described face, for calculating the fat or thin degree RF of face according to described face feature parameter; Described To Be Protected from Heat degree computing module, for the degree L that calculates according to the fat or thin degree RF of described face that To Be Protected from Heat; Described air conditioner operation parameters adjusting module, for producing the control instruction of adjustment air conditioner operation parameters according to described To Be Protected from Heat degree L.
Described face fat or thin degree computing module comprises and cuts the fat or thin degree computing unit of unit and face; Describedly cutting unit, for cutting described picture according to face feature parameter, obtaining face's picture of below eyes; The fat or thin degree computing unit of described face, for analyzing the picture after cutting, obtains fat or thin degree RF, the RF=skin pixels quantity/picture pixels quantity of face by the ratio calculating skin pixels quantity and picture pixels quantity.
Described To Be Protected from Heat degree computing module comprises expection body fat rate computing unit, every day rest energy expenditure computing unit and To Be Protected from Heat degree computing unit; Described expection body fat rate computing unit, converts fat or thin for face degree RF to expection body fat rate EBFR for the mode by data normalization; Described every day rest energy expenditure computing unit, for calculating rest energy expenditure RDEE every day according to expection body fat rate EBFR and average batheroom scale; Described To Be Protected from Heat degree computing unit, by rest energy expenditure RDEE every day is normalized between 0 to 1, the degree L that obtains that To Be Protected from Heat.
A kind of air-conditioning, comprises the system of carrying out airconditioning control based on face feature described in technique scheme.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (10)
1. carry out a method for airconditioning control based on face feature, it is characterized in that, comprise the steps:
Step 1, receives the picture that has face, carries out face feature analysis according to described picture, obtains face feature parameter;
Step 2, calculates the fat or thin degree RF of face according to described face feature parameter;
Step 3, according to the fat or thin degree RF of the described face degree L that calculates that To Be Protected from Heat;
Step 4, produces the control instruction of adjustment air conditioner operation parameters according to described To Be Protected from Heat degree L.
2. a kind of method of carrying out airconditioning control based on face feature according to claim 1, is characterized in that, being implemented as of step 2:
Step 2.1, selects at least one facial feature points, is calculated cut reference point by the facial feature points selected, and will cut the above picture excision of reference point, and retain residue face picture;
Step 2.2, analyzes the face's picture after cutting, and obtains fat or thin degree RF, the RF=skin pixels quantity/picture pixels quantity of face by the ratio calculating skin pixels quantity and picture pixels quantity.
3. a kind of method of carrying out airconditioning control based on face feature according to claim 2, it is characterized in that, the facial feature points of selection calculates with weighting or average mode and cuts reference point by described step 2.1; The facial feature points of described selection is the characteristic point of ocular vicinity.
4. a kind of method of carrying out airconditioning control based on face feature according to claim 1, is characterized in that, being implemented as of step 3:
Step 3.1, converts fat or thin for face degree RF to expection body fat rate EBFR by the mode of data normalization;
Step 3.2, calculates rest energy expenditure RDEE every day according to expection body fat rate EBFR and average batheroom scale;
Step 3.3, by rest energy expenditure RDEE every day is normalized between 0 to 1, the degree L that obtains that To Be Protected from Heat.
5. a kind of method of carrying out airconditioning control based on face feature according to claim 4, is characterized in that,
Expect described in step 3.1 that the computing formula of body fat rate EBFR is as follows:
EBFR＝(Rb-Ra)x[(RF-N)/(M-N)]+Ra；
Wherein, described Rb is the body fat rate upper limit, and Ra is body fat rate lower limit, and RF represents the fat or thin degree of face, and N represents the fat or thin degree lower limit of face, and M represents the fat or thin degree upper limit of face;
Described Rb, Ra are according to the presence or absence of gender data, and the gender data when determining sex is determined, have for:
Sex (women s=0, male sex s=1, countless according to s=2);
IfRF<NorRF>M, cannot from picture identification BFM;
If (s==0) Ra=women body fat rate lower limit, Rb=women's body fat rate higher limit;
ElseIf (s==1) Ra=male sex body fat rate lower limit, Rb=male sex's body fat rate higher limit;
ElseRa=gives tacit consent to body fat rate lower limit, and Rb=gives tacit consent to body fat rate higher limit;
Every day described in step 3.2, the computing formula of rest energy expenditure RDEE was as follows:
RDEE＝370+21.6x(1－EBFR)xKg；
Wherein, Kg is average weight, when differentiating age level and/or sex by picture, then calculates according to compatriots' average weight, if go out age level and/or sex by picture identifiable design, then calculates average weight according to this age level;
The computing formula of described in step 3.3, To Be Protected from Heat degree L is as follows:
IF(RDEE<Ea)L＝0；
ELSEIF(RDEE>Eb)L＝1；
ELSEL＝(RDEE-Ea)/(Eb-Ea)；
Wherein, Ea is the lower bound of rest energy expenditure RDEE every day, and Eb is the upper bound of rest energy expenditure RDEE every day;
Ea＝370+21.6x(1-Rb)xKa；Eb370+21.6x(1-Ra)xKb；
Wherein, Ka is body weight lower bound, and Kb is the body weight upper bound.
6. a kind of method of carrying out airconditioning control based on face feature according to claim 1, it is characterized in that, the specific implementation of step 4 comprises:
According to one or more in described To Be Protected from Heat degree adjustment air-conditioner temperature, humidity, air quantity, Auto On Time point/Auto Power On distance, automatic shutdown time point/automatic shutdown distance, continuous temperature change curve and User Interface, wherein said Auto Power On Distance geometry automatic shutdown distance all refers to the distance between user to air-conditioning equipment;
Also comprise and carry out colony's air-conditioning parameter optimization adjustment, be specially:
When the pictures received has multiple face, calculate each face respectively To Be Protected from Heat degree, by everyone, To Be Protected from Heat that degree is weighted or average computation, draws colony To Be Protected from Heat degree, in conjunction with colony's To Be Protected from Heat degree adjustment air conditioner operation parameters.
7. carry out a system for airconditioning control based on face feature, it is characterized in that, comprise face feature analysis module, the fat or thin degree computing module of face, To Be Protected from Heat degree computing module and air conditioner operation parameters adjusting module;
Described face feature analysis module, for receiving the picture that has face, carries out face feature analysis according to described picture, obtains face feature parameter;
The fat or thin degree computing module of described face, for calculating the fat or thin degree RF of face according to described face feature parameter;
Described To Be Protected from Heat degree computing module, for the degree L that calculates according to the fat or thin degree RF of described face that To Be Protected from Heat;
Described air conditioner operation parameters adjusting module, for producing the control instruction of adjustment air conditioner operation parameters according to described To Be Protected from Heat degree L.
8. a kind of system of carrying out airconditioning control based on face feature according to claim 7, is characterized in that, described face fat or thin degree computing module comprises and cuts the fat or thin degree computing unit of unit and face;
Describedly cutting unit, for cutting described picture according to face feature parameter, obtaining the face's picture cutting below reference point;
The fat or thin degree computing unit of described face, for analyzing the picture after cutting, obtains fat or thin degree RF, the RF=skin pixels quantity/picture pixels quantity of face by the ratio calculating skin pixels quantity and picture pixels quantity.
9. a kind of system of carrying out airconditioning control based on face feature according to claim 7, is characterized in that, described To Be Protected from Heat degree computing module comprises expection body fat rate computing unit, every day rest energy expenditure computing unit and To Be Protected from Heat degree computing unit;
Described expection body fat rate computing unit, converts fat or thin for face degree RF to expection body fat rate EBFR for the mode by data normalization;
Described every day rest energy expenditure computing unit, for calculating rest energy expenditure RDEE every day according to expection body fat rate EBFR and average batheroom scale;
Described To Be Protected from Heat degree computing unit, by rest energy expenditure RDEE every day is normalized between 0 to 1, the degree L that obtains that To Be Protected from Heat.
10. an air-conditioning, is characterized in that, comprises the system of based on face feature carrying out airconditioning control of claim requirement according to any one of 7-9.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510733881.2A CN105180385A (en) | 2015-10-30 | 2015-10-30 | Method and system for controlling air conditioner based on facial features and air conditioner |
| CN201610460314.9A CN106052034B (en) | 2015-10-30 | 2016-06-21 | A kind of method, system and air-conditioning carrying out airconditioning control based on face feature |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510733881.2A CN105180385A (en) | 2015-10-30 | 2015-10-30 | Method and system for controlling air conditioner based on facial features and air conditioner |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105180385A true CN105180385A (en) | 2015-12-23 |
Family
ID=54902652
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510733881.2A Withdrawn CN105180385A (en) | 2015-10-30 | 2015-10-30 | Method and system for controlling air conditioner based on facial features and air conditioner |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105180385A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106352506A (en) * | 2016-10-31 | 2017-01-25 | 广东美的制冷设备有限公司 | Air conditioner and automatic control method thereof |
| CN106382720A (en) * | 2016-08-31 | 2017-02-08 | 广东美的制冷设备有限公司 | Control method and system for air conditioner and air conditioner |
| CN106382719A (en) * | 2016-08-30 | 2017-02-08 | 广东美的制冷设备有限公司 | Air conditioner control method and air conditioner control system based on router |
| CN106403150A (en) * | 2016-08-31 | 2017-02-15 | 广东美的制冷设备有限公司 | Air conditioner and control method and system thereof |
| CN106837842A (en) * | 2017-02-28 | 2017-06-13 | 京东方科技集团股份有限公司 | A kind of method of intelligent fan and control intelligent fan |
| CN107940687A (en) * | 2017-11-15 | 2018-04-20 | 广东美的制冷设备有限公司 | Progress control method, operating control device and air conditioner |
| WO2019003639A1 (en) * | 2017-06-29 | 2019-01-03 | 三菱重工サーマルシステムズ株式会社 | Air conditioning control device, air conditioning system, and air conditioning control method and program |
| CN109196516A (en) * | 2016-06-03 | 2019-01-11 | 三菱电机株式会社 | Plant control unit and apparatus control method |
| CN109477655A (en) * | 2016-08-08 | 2019-03-15 | 三菱电机株式会社 | Air conditioner |
| TWI678091B (en) * | 2018-08-01 | 2019-11-21 | 台灣松下電器股份有限公司 | Internet of things home appliance system and control method thereof |
| CN111023510A (en) * | 2019-12-25 | 2020-04-17 | 天津商业大学 | Air conditioner control system based on human face temperature monitoring and human face image tracking |
-
2015
- 2015-10-30 CN CN201510733881.2A patent/CN105180385A/en not_active Withdrawn
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109196516A (en) * | 2016-06-03 | 2019-01-11 | 三菱电机株式会社 | Plant control unit and apparatus control method |
| CN109477655B (en) * | 2016-08-08 | 2021-04-20 | 三菱电机株式会社 | Air conditioner |
| CN109477655A (en) * | 2016-08-08 | 2019-03-15 | 三菱电机株式会社 | Air conditioner |
| CN106382719A (en) * | 2016-08-30 | 2017-02-08 | 广东美的制冷设备有限公司 | Air conditioner control method and air conditioner control system based on router |
| CN106382719B (en) * | 2016-08-30 | 2019-10-15 | 广东美的制冷设备有限公司 | A kind of air conditioning control method and system based on router realization |
| CN106382720B (en) * | 2016-08-31 | 2019-01-22 | 广东美的制冷设备有限公司 | Control method, system and the air conditioner of air conditioner |
| CN106403150A (en) * | 2016-08-31 | 2017-02-15 | 广东美的制冷设备有限公司 | Air conditioner and control method and system thereof |
| CN106382720A (en) * | 2016-08-31 | 2017-02-08 | 广东美的制冷设备有限公司 | Control method and system for air conditioner and air conditioner |
| CN106352506A (en) * | 2016-10-31 | 2017-01-25 | 广东美的制冷设备有限公司 | Air conditioner and automatic control method thereof |
| CN106837842B (en) * | 2017-02-28 | 2018-09-04 | 京东方科技集团股份有限公司 | A kind of method of intelligent fan and control intelligent fan |
| CN106837842A (en) * | 2017-02-28 | 2017-06-13 | 京东方科技集团股份有限公司 | A kind of method of intelligent fan and control intelligent fan |
| WO2019003639A1 (en) * | 2017-06-29 | 2019-01-03 | 三菱重工サーマルシステムズ株式会社 | Air conditioning control device, air conditioning system, and air conditioning control method and program |
| CN107940687A (en) * | 2017-11-15 | 2018-04-20 | 广东美的制冷设备有限公司 | Progress control method, operating control device and air conditioner |
| TWI678091B (en) * | 2018-08-01 | 2019-11-21 | 台灣松下電器股份有限公司 | Internet of things home appliance system and control method thereof |
| CN111023510A (en) * | 2019-12-25 | 2020-04-17 | 天津商业大学 | Air conditioner control system based on human face temperature monitoring and human face image tracking |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105180385A (en) | Method and system for controlling air conditioner based on facial features and air conditioner | |
| CN109114741B (en) | Air conditioner control method and device, storage medium and air conditioner | |
| JP6735323B2 (en) | Air conditioner control method | |
| JP6333486B2 (en) | Device control apparatus and device control method | |
| CN110513840A (en) | Temprature control method and system based on smart home operating system | |
| Cosma et al. | Using the contrast within a single face heat map to assess personal thermal comfort | |
| US20170123442A1 (en) | System and Method of Smart and Energy-Saving Environmental Control | |
| CN104848467A (en) | Control method of air conditioner and air conditioner | |
| CN113091273A (en) | Method and device for controlling air conditioner and air conditioner | |
| WO2023273653A1 (en) | Method and apparatus for controlling air conditioner, air conditioner, and storage medium | |
| CN113357779B (en) | Control method and device for air conditioning and household appliance | |
| CN109812936B (en) | Air conditioner and air conditioner air outlet temperature adjusting method and device thereof | |
| CN110726222B (en) | Air conditioner control method and device, storage medium and processor | |
| CN113251626B (en) | Rotating speed adjusting method and device for air deflector and air conditioner | |
| CN112361544A (en) | Hotel air conditioner control method and device and hotel air conditioner | |
| CN109442688A (en) | A kind of air conditioning control method, device, storage medium and air-conditioning | |
| JP6861366B2 (en) | Air conditioner | |
| WO2018209740A1 (en) | Method and device for controlling air conditioner | |
| CN113932425B (en) | Method and device for controlling air conditioner and air conditioner | |
| CN114811829A (en) | Body-sensing-temperature-based variable frequency control method and device, air conditioner and storage medium | |
| CN106052034B (en) | A kind of method, system and air-conditioning carrying out airconditioning control based on face feature | |
| WO2024113920A1 (en) | Air conditioner and control method therefor | |
| CN110887176A (en) | Control method for air conditioner and air conditioner | |
| CA2933799A1 (en) | System and method for automatically generating a facial remediation design and application protocol to address observable facial deviations | |
| CN110410975A (en) | Control method, device and the apparatus of air conditioning of the apparatus of air conditioning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C04 | Withdrawal of patent application after publication (patent law 2001) | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20151223 |