CN111932630A - Personnel-oriented air supply regulation and control method and device based on image recognition - Google Patents

Personnel-oriented air supply regulation and control method and device based on image recognition Download PDF

Info

Publication number
CN111932630A
CN111932630A CN202010707311.7A CN202010707311A CN111932630A CN 111932630 A CN111932630 A CN 111932630A CN 202010707311 A CN202010707311 A CN 202010707311A CN 111932630 A CN111932630 A CN 111932630A
Authority
CN
China
Prior art keywords
personnel
air supply
indoor personnel
key nodes
dimensional
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.)
Granted
Application number
CN202010707311.7A
Other languages
Chinese (zh)
Other versions
CN111932630B (en
Inventor
李先庭
王欢
赵家安
王贵锦
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.)
Tsinghua University
Original Assignee
Tsinghua University
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 Tsinghua University filed Critical Tsinghua University
Priority to CN202010707311.7A priority Critical patent/CN111932630B/en
Priority claimed from CN202010707311.7A external-priority patent/CN111932630B/en
Publication of CN111932630A publication Critical patent/CN111932630A/en
Application granted granted Critical
Publication of CN111932630B publication Critical patent/CN111932630B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/70Determining position or orientation of objects or cameras
    • G06T7/77Determining position or orientation of objects or cameras using statistical methods
    • FMECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24HEATING; RANGES; VENTILATING
    • F24FAIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00Control or safety arrangements
    • F24F11/62Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
    • F24F11/63Electronic processing
    • F24F11/64Electronic processing using pre-stored data
    • FMECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24HEATING; RANGES; VENTILATING
    • F24FAIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00Control or safety arrangements
    • F24F11/70Control systems characterised by their outputs; Constructional details thereof
    • F24F11/72Control systems characterised by their outputs; Constructional details thereof for controlling the supply of treated air, e.g. its pressure
    • F24F11/74Control systems characterised by their outputs; Constructional details thereof for controlling the supply of treated air, e.g. its pressure for controlling air flow rate or air velocity
    • F24F11/77Control systems characterised by their outputs; Constructional details thereof for controlling the supply of treated air, e.g. its pressure for controlling air flow rate or air velocity by controlling the speed of ventilators
    • FMECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24HEATING; RANGES; VENTILATING
    • F24FAIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00Control or safety arrangements
    • F24F11/70Control systems characterised by their outputs; Constructional details thereof
    • F24F11/72Control systems characterised by their outputs; Constructional details thereof for controlling the supply of treated air, e.g. its pressure
    • F24F11/79Control systems characterised by their outputs; Constructional details thereof for controlling the supply of treated air, e.g. its pressure for controlling the direction of the supplied air
    • FMECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24HEATING; RANGES; VENTILATING
    • F24FAIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00Control or safety arrangements
    • F24F11/89Arrangement or mounting of control or safety devices
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/25Fusion techniques
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00Three dimensional [3D] modelling, e.g. data description of 3D objects
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/70Determining position or orientation of objects or cameras
    • G06T7/73Determining position or orientation of objects or cameras using feature-based methods
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/40Extraction of image or video features
    • G06V10/44Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/103Static body considered as a whole, e.g. static pedestrian or occupant recognition
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02BCLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
    • Y02B30/00Energy efficient heating, ventilation or air conditioning [HVAC]
    • Y02B30/70Efficient control or regulation technologies, e.g. for control of refrigerant flow, motor or heating

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Chemical & Material Sciences (AREA)
  • Mechanical Engineering (AREA)
  • Combustion & Propulsion (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Signal Processing (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Biology (AREA)
  • Multimedia (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Computation (AREA)
  • Human Computer Interaction (AREA)
  • Computer Graphics (AREA)
  • Geometry (AREA)
  • Software Systems (AREA)
  • Fluid Mechanics (AREA)
  • Artificial Intelligence (AREA)
  • Fuzzy Systems (AREA)
  • Mathematical Physics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Air Conditioning Control Device (AREA)

Abstract

The embodiment of the invention provides a personnel-oriented air supply regulation and control method and a device based on image recognition, wherein the method comprises the following steps: acquiring real-time images of indoor personnel distribution under at least three visual angles; extracting human skeleton key nodes from indoor personnel distribution real-time images to obtain two-dimensional coordinates of the human skeleton key nodes at different visual angles; three-dimensional reconstruction and data fusion are carried out on two-dimensional coordinates of human skeleton key nodes under different visual angles based on epipolar constraint criteria, and position, posture and activity intensity information of indoor personnel in a three-dimensional space are obtained; and based on the position, posture and activity intensity information of indoor personnel in the three-dimensional space, the air supply mode is regulated and controlled in real time according to a preset regulation and control strategy. The embodiment of the invention can realize accurate positioning of personnel in a larger space, realize accurate regulation and control of an air supply mode according to the positioning of the personnel, complete guarantee or control of local space of the personnel and obviously reduce the energy consumption of an air conditioning system.

Description

Personnel-oriented air supply regulation and control method and device based on image recognition
Technical Field
The invention relates to the technical field of intelligent buildings, in particular to a personnel air supply-oriented regulation and control method and device based on image recognition.
Background
Air conditioning systems for large spaces usually consume excessive air volume and cold and heat to ensure the environment of the whole space, resulting in a large amount of energy consumption. The strategy of local guarantee is adopted by adopting the ventilation facing to the requirement, so that the requirement of air conditioner energy consumption can be obviously reduced.
However, there is a high requirement for an indoor personnel positioning system for personnel local environment guarantee or control, and currently, there are indoor personnel positioning modes such as infrared thermal imaging positioning, bluetooth positioning, and radio frequency positioning, but these positioning systems have certain defects in positioning speed, accuracy, convenience, and stability in practical application.
Disclosure of Invention
The embodiment of the invention provides a method and a device for regulating and controlling air supply to people based on image recognition, which overcome the problems or at least partially solve the problems.
In a first aspect, an embodiment of the present invention provides a method for regulating and controlling air supply to a person based on image recognition, including:
acquiring real-time images of indoor personnel distribution under at least three visual angles;
extracting human skeleton key nodes from the indoor personnel distribution real-time image to obtain two-dimensional coordinates of the human skeleton key nodes at different visual angles;
three-dimensional reconstruction and data fusion are carried out on the two-dimensional coordinates of the human body skeleton key nodes under different visual angles based on epipolar constraint criteria, and the position, the posture and the activity intensity information of indoor personnel in a three-dimensional space are obtained;
and adjusting and controlling the air supply mode in real time according to a preset adjusting and controlling strategy based on the position, posture and activity intensity information of the indoor personnel in the three-dimensional space.
Optionally, the data fusion and three-dimensional reconstruction are performed on the two-dimensional coordinates of the human skeleton key nodes under different viewing angles based on the epipolar constraint criterion, so as to obtain the position, posture and activity intensity information of indoor personnel in a three-dimensional space, and the method specifically includes:
carrying out epipolar constraint inspection on the two-dimensional coordinates of the human skeleton key nodes under different viewing angles by using camera calibration data to obtain polar surface geometric constraint residual errors among the human skeleton key nodes under different viewing angles;
screening out two-dimensional coordinates of key nodes of the human skeleton under two different visual angles with the polar surface geometric constraint residual error smaller than or equal to a preset threshold value based on the statistical data of the key positions of the human body, and taking the two-dimensional coordinates as a pair of matching estimation;
calculating horizontal distance and direction difference between two-dimensional coordinates of the key nodes of the human skeleton at the rest single visual angle, and performing multi-pair matching check based on the horizontal distance and the direction difference;
and performing multi-point repeated matching detection on the reconstructed overlapped points under the multiple visual angles, and performing single-point matching inspection on the matching results of the remaining single visual angles to obtain the position, posture and activity intensity information of the indoor personnel in the three-dimensional space.
Optionally, the regulation and control strategy aims at meeting the thermal comfort requirement of the personnel, and accordingly, based on the position, posture and activity intensity information of the indoor personnel in the three-dimensional space, the air supply mode is regulated and controlled in real time according to a preset regulation and control strategy, specifically:
adjusting the air inlet for air supply based on the position information of the indoor personnel in the three-dimensional space, wherein the adjustment comprises the opening of the air inlet and the adjustment of the air supply speed and direction of the opened air inlet;
and adjusting the temperature and speed of the air supply based on the posture information and the activity intensity information of the indoor personnel.
Optionally, the control strategy is aimed at controlling pollutants meeting the process cleanliness, and accordingly, based on the position, posture and activity intensity information of the indoor personnel in the three-dimensional space, the air supply mode is controlled in real time according to a preset control strategy, specifically:
adjusting the air inlet of the air supply based on the position information of the indoor personnel in the three-dimensional space, including opening the air inlet;
and adjusting the air supply speed based on the posture information and the activity intensity information of the indoor personnel.
In a second aspect, an embodiment of the present invention provides a regulation and control device for supplying air to a person based on image recognition, including:
the image acquisition module is used for acquiring indoor personnel distribution real-time images under at least three visual angles;
the framework node extraction module is used for extracting human framework key nodes from the indoor personnel distribution real-time image to obtain two-dimensional coordinates of the human framework key nodes at different visual angles;
the three-dimensional reconstruction module is used for performing three-dimensional reconstruction and data fusion on the two-dimensional coordinates of the human skeleton key nodes under different visual angles based on epipolar constraint criteria to obtain the position, posture and activity intensity information of indoor personnel in a three-dimensional space;
and the air supply regulation and control module is used for regulating and controlling the air supply mode in real time according to a preset regulation and control strategy based on the position, posture and activity intensity information of the indoor personnel in the three-dimensional space.
Optionally, the three-dimensional reconstruction module is specifically configured to:
carrying out epipolar constraint inspection on the two-dimensional coordinates of the human skeleton key nodes under different viewing angles by using camera calibration data to obtain polar surface geometric constraint residual errors among the human skeleton key nodes under different viewing angles;
screening out two-dimensional coordinates of key nodes of the human skeleton under two different visual angles with the polar surface geometric constraint residual error smaller than or equal to a preset threshold value based on the statistical data of the key positions of the human body, and taking the two-dimensional coordinates as a pair of matching estimation;
calculating horizontal distance and direction difference between two-dimensional coordinates of the key nodes of the human skeleton at the rest single visual angle, and performing multi-pair matching check based on the horizontal distance and the direction difference;
and performing multi-point repeated matching detection on the reconstructed overlapped points under the multiple visual angles, and performing single-point matching inspection on the matching results of the remaining single visual angles to obtain the position, posture and activity intensity information of the indoor personnel in the three-dimensional space.
Optionally, the regulation and control strategy aims at meeting a thermal comfort requirement of a person, and accordingly, the air supply regulation and control module is specifically configured to:
adjusting the air inlet for air supply based on the position information of the indoor personnel in the three-dimensional space, wherein the adjustment comprises the opening of the air inlet and the adjustment of the air supply speed and direction of the opened air inlet;
and adjusting the temperature and speed of the air supply based on the posture information and the activity intensity information of the indoor personnel.
Optionally, the control strategy targets pollutant control that meets process cleanliness, and accordingly, the air supply control module is specifically configured to:
adjusting the air inlet of the air supply based on the position information of the indoor personnel in the three-dimensional space, including opening the air inlet;
and adjusting the air supply speed based on the posture information and the activity intensity information of the indoor personnel.
In a third aspect, an embodiment of the present invention provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of the method for regulating and controlling air supply to a person based on image recognition according to the first aspect when executing the program.
In a fourth aspect, an embodiment of the present invention provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the method for regulating and controlling human-oriented air supply based on image recognition as provided in the first aspect.
The regulation and control method and the regulation and control device for the air supply facing the personnel based on the image recognition adopt a plurality of cameras to record and analyze the position information of the personnel from different angles, extract the key nodes of the personnel in the two-dimensional image, then, three-dimensional matching is carried out, a plurality of key nodes of personnel are adopted for carrying out matching calculation and screening, the position and body information of the personnel are distinguished by utilizing the matching results of the plurality of key nodes, the personnel in a larger space can be accurately positioned without contact and personnel cooperation, the defects of the existing positioning system in positioning speed, precision, convenience and stability can be overcome, and simultaneously according to the accurate position and body information of the personnel, an uneven environment is created in the indoor space, and the air supply mode is adjusted according to the position distribution of personnel in the indoor space, so that the energy consumption of the air conditioning system is reduced.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a regulation and control method for supplying air to a person based on image recognition according to an embodiment of the present invention;
fig. 2 is a schematic diagram of an arrangement of four wide-angle cameras in a room according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of the regulation of the thermal comfort requirement of a person according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of the control of contaminants to meet process cleanliness according to an embodiment of the present invention;
FIG. 5 is a top view of a plant provided in accordance with an embodiment of the present invention;
fig. 6 is a schematic structural diagram of a regulation and control device for supplying air to people based on image recognition according to an embodiment of the present invention;
fig. 7 is a schematic physical structure diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a schematic flow diagram of a regulation and control method for supplying air to a person based on image recognition according to an embodiment of the present invention, where the regulation and control method for supplying air to a person based on image recognition includes:
100, acquiring indoor personnel distribution real-time images under at least three visual angles;
specifically, the indoor personnel positioning method and the indoor personnel positioning system perform indoor personnel positioning by acquiring real-time images of indoor personnel distribution under at least three visual angles.
Because any two visual angles can record and reconstruct the position information of space personnel, but the difficulty is that how to determine that key points of human bodies in two images shot under different visual angles correspond to the same person, and the indoor personnel positioning based on the two visual angles is not accurate enough, therefore, the embodiment of the invention calculates and matches by collecting real-time images of indoor personnel distribution under at least three visual angles.
Fig. 2 is a schematic diagram of an arrangement of four wide-angle cameras in a room according to an embodiment of the present invention. In an embodiment, as shown in fig. 2, in consideration that people in a room are distributed randomly during use and mutual shielding between people is not beneficial to performing subsequent processing of people positioning, the collecting real-time images of indoor people distribution at least three viewing angles specifically includes: the real-time image of indoor personnel distribution under four visual angles is collected to four wide-angle cameras arranged at four corners of the top of a room and pointing to the center of the room, so that most of areas of the room can be shot by a plurality of cameras from different angles, and the problem that personnel are shielded from each other can be effectively solved.
101, extracting human skeleton key nodes from the indoor personnel distribution real-time image to obtain two-dimensional coordinates of the human skeleton key nodes at different visual angles;
specifically, the human skeleton key node extraction algorithm is utilized to extract the human skeleton key nodes from the acquired indoor personnel distribution real-time image, and two-dimensional coordinates of the human skeleton key nodes under different visual angles are obtained, so that picture information of the human skeleton key nodes under different visual angles is converted into semantic information in a camera two-dimensional plane.
It should be noted that the human skeleton key nodes specifically include which nodes are related to the adopted extraction algorithm. In one embodiment, the human skeletal key nodes include: a human body left shoulder node, a human body right shoulder node and a human body neck node.
102, performing three-dimensional reconstruction and data fusion on the two-dimensional coordinates of the human skeleton key nodes under different visual angles based on epipolar constraint criteria to obtain the position, posture and activity intensity information of indoor personnel in a three-dimensional space;
since the distribution of the personnel in the space may be overlapped and the camera projection relationship, the spatial position and the body-related information of the personnel in the room need to be further extracted and confirmed through three-dimensional reconstruction and data fusion, wherein the body-related information refers to the posture and activity intensity information of the personnel. Three-dimensional registration and data fusion comprise a series of matching algorithms and inspection steps.
The method has the advantages that three-dimensional reconstruction and data fusion are carried out on the two-dimensional coordinates of the human skeleton key nodes under different visual angles based on the epipolar constraint criterion, the positions, postures and activity intensity of personnel can be rapidly, accurately and conveniently recognized, and the recognition task can be completed under the condition that the positions of the personnel are overlapped or other heat sources interfere.
And 103, regulating and controlling the air supply mode in real time according to a preset regulation and control strategy based on the position, posture and activity intensity information of the indoor personnel in the three-dimensional space.
On the basis of obtaining the position, posture and activity intensity information of indoor personnel in a three-dimensional space based on the image recognition and positioning technology, the air supply mode of the air conditioning system can be regulated and controlled in real time according to a preset regulation and control strategy, and the local environment of the personnel can be guaranteed or controlled.
For example, the distribution situation of indoor personnel can be determined according to the position information of the indoor personnel in the three-dimensional space, then the air conditioning system is switched into a corresponding air supply mode according to the acquired distribution situation of the indoor personnel, and in each mode, when the position and the posture of the personnel change, accurate real-time regulation and control are carried out according to a preset control strategy. Taking an air conditioning system meeting the thermal comfort of personnel as an example, when the position of the personnel changes in real time, an air conditioner within the range of R away from the personnel is started, and in summer, when the activity intensity of the personnel is low, cold air with higher temperature can be sent; in winter, when the activity intensity of the personnel is high, hot air with lower temperature can be sent. Taking an air conditioning system for controlling the type of the pollutants of the personnel as an example, when the position of the personnel changes in real time, the air conditioner within the range of R' away from the personnel is started, and when the activity intensity of the personnel is low, the air supply speed is reduced.
The regulation and control method for personnel-oriented air supply based on image recognition provided by the embodiment of the invention adopts a plurality of cameras to record and analyze the position information of personnel from different angles, extracts key nodes of the personnel in a two-dimensional image, then, three-dimensional matching is carried out, a plurality of key nodes of one person are adopted for carrying out matching calculation and screening, the position and body information of the person are judged by utilizing the matching results of the plurality of key nodes, the accurate positioning of the person in a larger space can be realized in a non-contact way without the cooperation of the person, the defects of the existing positioning system in the aspects of positioning speed, precision, convenience and stability can be solved, and simultaneously according to the accurate position and body information of the person, an uneven environment is created in the indoor space, and the air supply mode is adjusted according to the position distribution of personnel in the indoor space, so that the energy consumption of the air conditioning system is reduced.
Further, on the basis of the embodiment shown in fig. 1, the step 102 specifically includes:
step 1021, performing epipolar constraint inspection on the two-dimensional coordinates of the human skeleton key nodes under different viewing angles by using camera calibration data to obtain polar surface geometric constraint residuals among the human skeleton key nodes under different viewing angles;
specifically, the epipolar constraint is to check, namely calculate a group of polar plane geometric constraint residual errors of the human skeleton key nodes in the camera by using the camera calibration data, and screen out two-dimensional coordinates of the human skeleton key nodes of which the polar plane geometric constraint residual errors are less than or equal to an epipolar constraint threshold value.
The polar constraint check of the two-dimensional coordinates of the human skeleton key nodes under different visual angles by using the camera calibration data specifically comprises the steps of calculating polar surface geometric constraint residual errors of the human skeleton key nodes under any two different visual angles by using the camera calibration data, and screening out the two-dimensional coordinates of the human skeleton key nodes under two different visual angles, wherein the polar surface geometric constraint residual errors are less than or equal to a polar constraint threshold value.
It should be noted that, because the key nodes of the human skeleton are extracted based on image features, the extracted results of the images may be slightly different at different viewing angles, and therefore, a larger threshold is usually adopted for extreme constraint inspection to ensure that all potential matching objects can enter into subsequent screening.
In one embodiment, the step of calculating the polar geometry constraint residuals of the human skeleton key nodes at two different viewing angles comprises:
the residuals of the normalized pixel coordinates acquired by the different cameras are first calculated using the following equation (1), where,
Figure BDA0002595273040000081
wherein (x)1,y1) Is the two-dimensional coordinate of a key node of a human skeleton under a first camera (x)2,y2) Is a two-dimensional coordinate, s, of a key node of a human skeleton under a second camera1,2Residual error of normalized pixel coordinates of the human skeleton key node, F, acquired for different cameras12From camera calibration data;
then based on the residual error of the normalized pixel coordinate acquired by different cameras, the following formula (2) is utilized to calculate and obtain the polar surface geometric constraint residual error of the human skeleton key node under two different visual angles,
Figure BDA0002595273040000091
wherein s isp1,p2The polar surface geometric constraint residual errors of the key nodes of the human skeleton under two different visual angles,
Figure BDA0002595273040000092
the residuals of the normalized pixel coordinates of the human skeleton key node L acquired by p2 for two different cameras p1,
Figure BDA0002595273040000093
the residuals of the normalized pixel coordinates of the human skeleton key node R acquired by p2 for two different cameras p1,
Figure BDA0002595273040000094
the residual error of the normalized pixel coordinates of the human skeleton key node N acquired by p2 for two different cameras p1, L represents the human left shoulder node, R represents the human right shoulder node, and N represents the human neck node.
Step 1022, screening out two-dimensional coordinates of the key nodes of the human skeleton under two different viewing angles with the polar surface geometric constraint residual error less than or equal to a preset threshold value based on the statistical data of the key positions of the human body, and taking the two-dimensional coordinates as a pair of matching estimation;
because the threshold value of the epipolar constraint is large, other constraint relations are also needed to check the matching reasonability, and the embodiment of the invention is mainly developed aiming at the reasonability of data and the uniqueness of matching.
The rationality of the match is first checked by the rationality of the data. Considering that each part of a human body is difficult to be shot in one picture due to the posture relation, two-dimensional coordinates of key nodes of the human body skeleton under two different visual angles with the polar surface geometric constraint residual error smaller than or equal to a preset threshold value are screened out as a pair of matching estimation based on the statistical data of the key positions of the human body, so that the identification and matching analysis of the key positions of the human body are realized.
The human body key position statistical data is the statistical data which can be directly obtained and is related to each key position of the human body. In one embodiment, the human body key positions include: left and right shoulders and neck. Namely, the identification and matching analysis is performed only for the positions of the left and right shoulders and neck, including shoulder height check, shoulder width check, etc. The shoulder height detection is carried out according to the position of the shoulder of a person in a room, and the matching judgment is carried out, in one embodiment, the value range of the shoulder height detection is 0.2-2m, and the shoulder height detection is considered to be wrong matching when the value range exceeds the range; the shoulder width check is to screen out wrong matches according to the range of the human shoulder width, the value range is 0.3-0.5m, and matches beyond the range are deleted.
Step 1023, calculating horizontal distance and direction difference between two-dimensional coordinates of the human skeleton key nodes of the remaining single visual angle, and performing multi-pair matching check based on the horizontal distance and the direction difference;
further, the match is checked for uniqueness. The uniqueness of matching means that a group of key node data of a person in a certain camera can only be matched with a group of key node data of other cameras, and the matching cannot be repeated.
In the embodiment of the invention, a clustering algorithm based on distance is utilized to carry out multi-pair matching check.
Specifically, the horizontal distance between two estimates is calculated according to formula (3), and the direction difference between the two estimates is calculated according to formula (4), where the estimate refers to the two-dimensional coordinates of the human skeleton key nodes at a remaining certain viewing angle, and the two-dimensional coordinates of the human skeleton key nodes at the remaining single viewing angle refer to the two-dimensional coordinates of the human skeleton key nodes which are not successfully matched and remain after step 1022. Wherein,
Figure BDA0002595273040000101
wherein,
Figure BDA0002595273040000102
is the two-dimensional coordinates of the human skeleton key node N under the first camera p1,
Figure BDA0002595273040000103
is the two-dimensional coordinate of the key node N of the human skeleton under the first camera p2, Dp1,p2N represents a human neck node for the horizontal distance between the two estimates;
Figure BDA0002595273040000104
wherein,
Figure BDA0002595273040000105
is the two-dimensional coordinates of the human skeleton key node L under the first camera p1,
Figure BDA0002595273040000106
is the two-dimensional coordinate of the key node L of the human skeleton under the second camera p2,
Figure BDA0002595273040000107
is the two-dimensional coordinates of the human skeleton key node R under the first camera p1,
Figure BDA0002595273040000108
is a key joint of the human skeleton under a second camera p2Two-dimensional coordinates of point R, Ap1,p2For the directional difference between the two estimates, L represents the human left shoulder node and R represents the human right shoulder node.
And then, comparing the horizontal distance with a preset horizontal distance threshold, comparing the direction difference with a preset direction difference threshold, and judging as homologous when the horizontal distance is less than or equal to the preset horizontal distance threshold and the direction difference is less than or equal to the preset direction difference threshold. In one embodiment, the same person is determined to be from when the distance between the two estimates is less than 0.3m and the direction difference is less than 45 °.
And 1024, performing multi-point repeated matching detection on the reconstructed overlapped points under the multiple visual angles, and performing single-point matching inspection on the matching results of the remaining single visual angles to obtain the position, posture and activity intensity information of the indoor personnel in the three-dimensional space.
In order to further improve the accuracy of indoor personnel positioning, the embodiment of the invention also carries out multi-point repeated matching detection on the reconstructed overlapped points under multiple visual angles, wherein the repeated matching is to check the reconstructed overlapped points of multiple cameras, judge whether each matching meets the uniqueness or not and then store the confirmed result. In order to take the result of the human body key points shot by only one group of cameras into account, the embodiment of the invention also utilizes the confirmed matching to check the matching uniqueness of the matching results of the remaining single group of cameras (namely, single view angle), and the single group of matching results after screening and confirmation are also included in the qualified matching.
According to the regulation and control method for personnel air supply based on image recognition, provided by the embodiment of the invention, the position information of personnel is recorded and analyzed from different angles by adopting a plurality of cameras, the key nodes of the personnel in a two-dimensional image are extracted, then three-dimensional matching is carried out, the matching calculation and screening are carried out by adopting a plurality of key nodes of one personnel, the judgment of the position and body information of the personnel is realized by utilizing the matching results of the plurality of key nodes, and the accurate positioning of the personnel in a larger space can be realized in a non-contact manner without the cooperation of the personnel.
Based on the content of the above embodiments, the regulation and control strategy aims at meeting the thermal comfort requirement of personnel, or aims at meeting the pollutant control of process cleanliness.
Specifically, the acquired information of the position, the posture and the activity intensity of the indoor personnel in the three-dimensional space is transmitted to the data processing and control module of the air conditioning system, and then the data processing and control module can regulate and control the air supply mode of the air conditioning system in real time according to a preset regulation and control strategy.
Further, in a specific mode, when the position of indoor personnel changes, the positioning system transmits the real-time position, posture and activity intensity information of the indoor personnel in a three-dimensional space to the data processing and control module, and the data processing and control module adjusts air supply parameters of each air supply outlet in real time according to a corresponding regulation and control strategy, so that the local environment of the personnel is guaranteed or controlled. Through the combination of design and operation regulation and control, ventilation facing to personnel requirements can be realized more finely.
The air conditioner of large space using personnel as main service object or control object through more reasonable supply-air outlet system design to and according to the accurate regulation and control of personnel's real-time position and relevant information of health, more efficient guarantees or control personnel's local environment simultaneously, can realize showing energy-conserving effect.
It can be understood that the regulation and control method for personnel-oriented air supply based on image recognition provided by the embodiment of the invention can meet the operation regulation of an air conditioning system mainly based on thermal comfort of personnel, can also realize the design operation regulation and control of the air conditioning system mainly based on pollutant control of personnel, and can realize the reduction of air volume and the reduction of energy consumption of the air conditioning system under the condition of effectively controlling pollutants.
Situations where regulatory strategies are aimed at meeting personnel thermal comfort needs
The ventilation facing the requirement does not need to carry out environment control on all the spaces, and the corresponding air supply outlet is adjusted by identifying the position of personnel, so that the environment protection of the local range of the personnel is realized, and the energy consumption of the air conditioner can be obviously reduced. For air supply of an air conditioner, air supply from an air supply outlet close to a person is more efficient. Under the condition of meeting the comfort of personnel, the close air supply outlet can meet the requirement of the personnel by lower air quantity or less cold and heat quantity.
Further, under the condition that the regulation and control strategy aims at meeting the thermal comfort requirement of personnel, the air supply mode is regulated and controlled in real time according to a preset regulation and control strategy based on the position, posture and activity intensity information of the indoor personnel in the three-dimensional space, and the regulation and control strategy specifically comprises the following steps:
adjusting the air inlet for air supply based on the position information of the indoor personnel in the three-dimensional space, wherein the adjustment comprises the opening of the air inlet and the adjustment of the air supply speed and direction of the opened air inlet;
and adjusting the temperature and speed of the air supply based on the posture information and the activity intensity information of the indoor personnel.
Fig. 3 is a schematic diagram of the regulation aiming at satisfying the thermal comfort requirement of the person according to the embodiment of the present invention.
As shown in fig. 3, there are a plurality of areas in a large space, a process a area, a process B area, and a C area for transferring goods, wherein the A, B area is an area where personnel work, but the working time and the number of the personnel are different.
According to two typical working scenes, the process A operates independently, the process B operates independently, two sets of air supply modes are designed, and air supply outlets are arranged in the area where people are located. And designing the air conditioning system for the two typical scenes respectively.
The method comprises the steps of obtaining indoor personnel distribution real-time images under at least three visual angles in the large space, carrying out human skeleton key node extraction on the indoor personnel distribution real-time images, obtaining two-dimensional coordinates of human skeleton key nodes under different visual angles, carrying out three-dimensional reconstruction and data fusion on the two-dimensional coordinates of the human skeleton key nodes under different visual angles based on antipodal constraint criteria, obtaining position, posture and activity intensity information of indoor personnel in a three-dimensional space, further identifying a working scene, and starting a corresponding air supply mode, for example, starting air supply outlets in a R meter range near the personnel (R value is related to the arrangement condition of the air outlets, and the range at least comprises one air supply outlet).
Further, the activity intensity of the person significantly affects the thermal comfort of the human body, and the activity intensity information of the human body can be recognized by the image recognition technology, and the activity intensity information in the activity state is classified into mild (sitting still, standing still, etc.), moderate (having a certain activity), and severe (carrying, etc.). When the personnel move in the air supply mode, the air supply outlets are independently and accurately regulated, for example, the air supply speed of the air supply outlet close to the personnel is controlled to be 1m/s, the air supply of the other air supply outlets is closed, and when the quantity of the personnel is reduced, the air supply temperature can be properly increased. When personnel move, the air ports are switched in real time, and the local environment of the personnel is accurately guaranteed.
For a room cooled in summer, the ambient temperature is high, and cold air needs to be supplied to people. The smaller the activity intensity of the human body is, the smaller the heat quantity is released, and the less the cold quantity provided by the air supply of the air conditioner is needed to meet the same thermal comfort level, so that the lower the activity intensity is, the cold air with higher temperature can be supplied in summer, and the purpose of energy conservation is achieved.
For a room heated in winter, the ambient temperature is low, and hot air needs to be provided for personnel. The greater the activity intensity of the human body is, the greater the heat released is, the less the heat provided by the air supply of the air conditioner is required to meet the same thermal comfort level, therefore, the higher the activity intensity is, the hot air with lower temperature can be supplied in winter, and the purpose of energy saving is achieved.
The real-time personnel position is obtained through a non-contact personnel positioning system, the control of air supply is realized according to a corresponding regulation and control strategy, the local environment is controlled, and the thermal comfort of personnel is realized with less energy consumption.
(II) case of controlling pollutant to satisfy process cleanliness
Due to the breathing of human body and the shedding of skin cells, a large amount of particles are generated, which can affect some production processes (such as the production of electronic chips), so that the particles need to be controlled. Similarly, the particulate matter that personnel produced can be got rid of to the wind gap that is close to personnel more efficient, therefore the wind gap that is close to personnel can realize the same cleanliness factor level with less amount of wind, realizes energy-conservation.
Use the electron clean room as the example, the electron clean room need maintain the cleanliness factor in the clean room through a large amount of filtration air supplies, and to high-grade electron clean room, the personnel are main dust production source, but a large amount of air supply units are arranged to traditional electron clean room top, can not adjust according to personnel's condition, move under the design condition all the time, have higher energy consumption. But the position of the personnel is identified, and the air quantity of the air supply outlet is adjusted, so that the remarkable energy saving is realized under the same environment protection effect.
Further, based on the position, posture and activity intensity information of the indoor personnel in the three-dimensional space, the air supply mode is regulated and controlled in real time according to a preset regulation and control strategy, and the method specifically comprises the following steps:
adjusting the air inlet of the air supply based on the position information of the indoor personnel in the three-dimensional space, including opening the air inlet;
and adjusting the air supply speed based on the posture information and the activity intensity information of the indoor personnel.
Personnel's activity intensity can show the level that influences personnel and produce the particulate matter, and when activity intensity is high, the quantity that the particulate matter produced is very showing and is rising, needs more amount of wind, consequently based on indoor personnel's attitude information and activity intensity information judge personnel activity intensity, can reduce the air supply when activity intensity is weak. Significant energy saving can be achieved through local environment protection based on the position of the person and the activity intensity.
Fig. 4 is a schematic control diagram for controlling contaminants to meet process cleanliness according to an embodiment of the present invention. As shown in fig. 4, the electronic factory building has stable production conditions, and the air supply design is a mode, here, for example, a thousand-level clean room, FFUs are uniformly arranged on the top of the factory building according to an arrangement rate of 25%, and the air supply speed is 0.35m/s in the design state, and fig. 5 is a top view of the factory building provided by the embodiment of the present invention. The factory building installer positioning system acquires real-time indoor personnel distribution images under at least three visual angles through a plurality of high-definition cameras; extracting human skeleton key nodes from the indoor personnel distribution real-time image to obtain two-dimensional coordinates of the human skeleton key nodes at different visual angles; three-dimensional reconstruction and data fusion are carried out on the two-dimensional coordinates of the human body skeleton key nodes under different visual angles based on epipolar constraint criteria, and the position, the posture and the activity intensity information of indoor personnel in a three-dimensional space are obtained; the method comprises the steps of transmitting the position, the posture and the activity intensity information of indoor personnel in a three-dimensional space to a data processing and control module, and carrying out real-time regulation and control on an air supply mode according to a preset regulation and control strategy on the basis of the position, the posture and the activity intensity information of the indoor personnel in the three-dimensional space. In one embodiment, the specific regulatory strategy is as follows:
(1) when the person is in position P1 in fig. 5, the FFU speed directly above the adjustment person is 0.35m/s with light activity and the remaining FUUs are turned off.
(2) When the person is in position P2 in fig. 5, the FFU speed on both sides of the person is adjusted to 0.35m/s with light activity, and the remaining FUUs are turned off.
(3) When the person is in position P3 in fig. 5, the activity is mild, the four FFUs closest to the person are adjusted at a speed of 0.35m/s, and the remaining FUUs are turned off.
The personnel-oriented air supply regulation and control method based on image recognition can conveniently, quickly and accurately realize air supply mode switching of the air conditioner in large space under different personnel distribution states and accurate air supply regulation and control during position change of personnel in each mode, complete local space guarantee or control of the personnel, and remarkably reduce energy consumption of the air conditioning system compared with the traditional full-space guaranteed air conditioning system.
Fig. 6 is a schematic structural diagram of a regulation and control device for supplying air to people based on image recognition, which includes: an image acquisition module 610, a skeleton node extraction module 620, a three-dimensional reconstruction module 630 and an air supply regulation and control module 640, wherein,
the image acquisition module 610 is used for acquiring indoor personnel distribution real-time images under at least three visual angles;
a skeleton node extraction module 620, configured to perform human skeleton key node extraction on the indoor personnel distribution real-time image, and obtain two-dimensional coordinates of human skeleton key nodes at different viewing angles;
the three-dimensional reconstruction module 630 is configured to perform three-dimensional reconstruction and data fusion on the two-dimensional coordinates of the human skeleton key nodes at different viewing angles based on epipolar constraint criteria, so as to obtain position, posture and activity intensity information of indoor personnel in a three-dimensional space;
and the air supply regulation and control module 640 is used for regulating and controlling the air supply mode in real time according to a preset regulation and control strategy based on the position, posture and activity intensity information of the indoor personnel in the three-dimensional space.
In specific implementation, the image acquisition module 610, the skeleton node extraction module 620 and the three-dimensional reconstruction module 630 form an indoor person positioning device, acquire position, posture and activity intensity information of indoor persons in a three-dimensional space, and send the position, posture and activity intensity information of the indoor persons in the three-dimensional space to the data processing and control module (i.e., the air supply regulation and control module 640);
the data processing and control module (i.e. the air supply regulation and control module 640) regulates and controls the air supply mode in real time according to the position, posture and activity intensity information of the indoor personnel in the three-dimensional space and a preset regulation and control strategy so as to realize local environment guarantee or pollutant control of the area where the indoor personnel are located.
If the accurate position information of the personnel can be obtained, a non-uniform environment can be created in the indoor space and the position distribution of the personnel in the indoor space can be adjusted, so that the energy consumption is greatly saved.
The regulation and control device for supplying air to people based on image recognition provided by the embodiment of the invention is used for realizing the regulation and control method embodiment for supplying air to people based on image recognition, therefore, the description in the method embodiment can be used for understanding each functional module in the device embodiment of the invention, and the description is not repeated.
The regulation and control device for personnel air supply based on image recognition provided by the embodiment of the invention adopts a plurality of cameras to record and analyze the position information of personnel from different angles, extracts key nodes of the personnel in a two-dimensional image, then, three-dimensional matching is carried out, a plurality of key nodes of personnel are adopted for carrying out matching calculation and screening, the position and body information of the personnel are distinguished by utilizing the matching results of the plurality of key nodes, the personnel in a larger space can be accurately positioned without contact and personnel cooperation, the defects of the existing positioning system in positioning speed, precision, convenience and stability can be overcome, and simultaneously according to the accurate position and body information of the personnel, an uneven environment is created in the indoor space, and the air supply mode is adjusted according to the position distribution of personnel in the indoor space, so that the energy consumption of the air conditioning system is reduced.
Based on the content of the foregoing embodiment, the three-dimensional reconstruction module 630 is specifically configured to:
carrying out epipolar constraint inspection on the two-dimensional coordinates of the human skeleton key nodes under different viewing angles by using camera calibration data to obtain polar surface geometric constraint residual errors among the human skeleton key nodes under different viewing angles;
screening out two-dimensional coordinates of key nodes of the human skeleton under two different visual angles with the polar surface geometric constraint residual error smaller than or equal to a preset threshold value based on the statistical data of the key positions of the human body, and taking the two-dimensional coordinates as a pair of matching estimation;
calculating horizontal distance and direction difference between two-dimensional coordinates of the key nodes of the human skeleton at the rest single visual angle, and performing multi-pair matching check based on the horizontal distance and the direction difference;
and performing multi-point repeated matching detection on the reconstructed overlapped points under the multiple visual angles, and performing single-point matching inspection on the matching results of the remaining single visual angles to obtain the position, posture and activity intensity information of the indoor personnel in the three-dimensional space.
Based on the content of the above embodiments, the regulation and control strategy aims to meet the thermal comfort requirement of the personnel, and accordingly, the air supply regulation and control module 640 is specifically configured to:
adjusting the air inlet for air supply based on the position information of the indoor personnel in the three-dimensional space, wherein the adjustment comprises the opening of the air inlet and the adjustment of the air supply speed and direction of the opened air inlet;
and adjusting the temperature and speed of the air supply based on the posture information and the activity intensity information of the indoor personnel.
Based on the content of the above embodiment, the regulation and control strategy aims at controlling the pollutants meeting the process cleanliness, and accordingly, the air supply regulation and control module 640 is specifically configured to:
adjusting the air inlet of the air supply based on the position information of the indoor personnel in the three-dimensional space, including opening the air inlet;
and adjusting the air supply speed based on the posture information and the activity intensity information of the indoor personnel.
Fig. 7 is a schematic entity structure diagram of an electronic device according to an embodiment of the present invention, and as shown in fig. 7, the electronic device may include: a processor (processor)710, a communication Interface (Communications Interface)720, a memory (memory)730, and a communication bus 740, wherein the processor 710, the communication Interface 720, and the memory 730 communicate with each other via the communication bus 740. The processor 710 may call a computer program stored on the memory 730 and executable on the processor 710 to perform the method for regulating and controlling the supply of air to people based on image recognition provided by the embodiments of the method, for example, including: acquiring real-time images of indoor personnel distribution under at least three visual angles; extracting human skeleton key nodes from the indoor personnel distribution real-time image to obtain two-dimensional coordinates of the human skeleton key nodes at different visual angles; three-dimensional reconstruction and data fusion are carried out on the two-dimensional coordinates of the human body skeleton key nodes under different visual angles based on epipolar constraint criteria, and the position, the posture and the activity intensity information of indoor personnel in a three-dimensional space are obtained; and adjusting and controlling the air supply mode in real time according to a preset adjusting and controlling strategy based on the position, posture and activity intensity information of the indoor personnel in the three-dimensional space.
In addition, the logic instructions in the memory 730 can be implemented in the form of software functional units and stored in a computer readable storage medium when the software functional units are sold or used as independent products. Based on such understanding, the technical solutions of the embodiments of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The embodiment of the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the method for regulating and controlling air supply to a person based on image recognition, which includes: acquiring real-time images of indoor personnel distribution under at least three visual angles; extracting human skeleton key nodes from the indoor personnel distribution real-time image to obtain two-dimensional coordinates of the human skeleton key nodes at different visual angles; three-dimensional reconstruction and data fusion are carried out on the two-dimensional coordinates of the human body skeleton key nodes under different visual angles based on epipolar constraint criteria, and the position, the posture and the activity intensity information of indoor personnel in a three-dimensional space are obtained; and adjusting and controlling the air supply mode in real time according to a preset adjusting and controlling strategy based on the position, posture and activity intensity information of the indoor personnel in the three-dimensional space.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims (10)

1. A regulation and control method for personnel-oriented air supply based on image recognition is characterized by comprising the following steps:
acquiring real-time images of indoor personnel distribution under at least three visual angles;
extracting human skeleton key nodes from the indoor personnel distribution real-time image to obtain two-dimensional coordinates of the human skeleton key nodes at different visual angles;
three-dimensional reconstruction and data fusion are carried out on the two-dimensional coordinates of the human body skeleton key nodes under different visual angles based on epipolar constraint criteria, and the position, the posture and the activity intensity information of indoor personnel in a three-dimensional space are obtained;
and adjusting and controlling the air supply mode in real time according to a preset adjusting and controlling strategy based on the position, posture and activity intensity information of the indoor personnel in the three-dimensional space.
2. The method for regulating and controlling personnel-oriented air supply based on image recognition according to claim 1, wherein the method for performing data fusion and three-dimensional reconstruction on the two-dimensional coordinates of the human skeleton key nodes under different viewing angles based on epipolar constraint criteria to obtain the position, posture and activity intensity information of indoor personnel in a three-dimensional space specifically comprises the following steps:
carrying out epipolar constraint inspection on the two-dimensional coordinates of the human skeleton key nodes under different viewing angles by using camera calibration data to obtain polar surface geometric constraint residual errors among the human skeleton key nodes under different viewing angles;
screening out two-dimensional coordinates of key nodes of the human skeleton under two different visual angles with the polar surface geometric constraint residual error smaller than or equal to a preset threshold value based on the statistical data of the key positions of the human body, and taking the two-dimensional coordinates as a pair of matching estimation;
calculating horizontal distance and direction difference between two-dimensional coordinates of the key nodes of the human skeleton at the rest single visual angle, and performing multi-pair matching check based on the horizontal distance and the direction difference;
and performing multi-point repeated matching detection on the reconstructed overlapped points under the multiple visual angles, and performing single-point matching inspection on the matching results of the remaining single visual angles to obtain the position, posture and activity intensity information of the indoor personnel in the three-dimensional space.
3. The method for regulating and controlling personnel-oriented air supply based on image recognition as claimed in claim 1, wherein the regulation and control strategy is aimed at meeting the thermal comfort requirement of personnel, and correspondingly, based on the position, posture and activity intensity information of the indoor personnel in the three-dimensional space, the air supply mode is regulated and controlled in real time according to a preset regulation and control strategy, specifically:
adjusting the air inlet for air supply based on the position information of the indoor personnel in the three-dimensional space, wherein the adjustment comprises the opening of the air inlet and the adjustment of the air supply speed and direction of the opened air inlet;
and adjusting the temperature and speed of the air supply based on the posture information and the activity intensity information of the indoor personnel.
4. The personnel-oriented air supply regulation and control method based on image recognition as claimed in claim 1, wherein the regulation and control strategy aims at controlling pollutants meeting process cleanliness, and accordingly, based on the position, posture and activity intensity information of the indoor personnel in the three-dimensional space, the air supply mode is regulated and controlled in real time according to a preset regulation and control strategy, specifically:
adjusting the air inlet of the air supply based on the position information of the indoor personnel in the three-dimensional space, including opening the air inlet;
and adjusting the air supply speed based on the posture information and the activity intensity information of the indoor personnel.
5. The utility model provides a towards regulation and control device of personnel's air supply based on image recognition which characterized in that includes:
the image acquisition module is used for acquiring indoor personnel distribution real-time images under at least three visual angles;
the framework node extraction module is used for extracting human framework key nodes from the indoor personnel distribution real-time image to obtain two-dimensional coordinates of the human framework key nodes at different visual angles;
the three-dimensional reconstruction module is used for performing three-dimensional reconstruction and data fusion on the two-dimensional coordinates of the human skeleton key nodes under different visual angles based on epipolar constraint criteria to obtain the position, posture and activity intensity information of indoor personnel in a three-dimensional space;
and the air supply regulation and control module is used for regulating and controlling the air supply mode in real time according to a preset regulation and control strategy based on the position, posture and activity intensity information of the indoor personnel in the three-dimensional space.
6. The device for regulating and controlling air supply to people based on image recognition as claimed in claim 5, wherein the three-dimensional reconstruction module is specifically configured to:
carrying out epipolar constraint inspection on the two-dimensional coordinates of the human skeleton key nodes under different viewing angles by using camera calibration data to obtain polar surface geometric constraint residual errors among the human skeleton key nodes under different viewing angles;
screening out two-dimensional coordinates of key nodes of the human skeleton under two different visual angles with the polar surface geometric constraint residual error smaller than or equal to a preset threshold value based on the statistical data of the key positions of the human body, and taking the two-dimensional coordinates as a pair of matching estimation;
calculating horizontal distance and direction difference between two-dimensional coordinates of the key nodes of the human skeleton at the rest single visual angle, and performing multi-pair matching check based on the horizontal distance and the direction difference;
and performing multi-point repeated matching detection on the reconstructed overlapped points under the multiple visual angles, and performing single-point matching inspection on the matching results of the remaining single visual angles to obtain the position, posture and activity intensity information of the indoor personnel in the three-dimensional space.
7. The device as claimed in claim 5, wherein the control strategy is aimed at meeting the thermal comfort requirement of the person, and accordingly the air supply control module is specifically configured to:
adjusting the air inlet for air supply based on the position information of the indoor personnel in the three-dimensional space, wherein the adjustment comprises the opening of the air inlet and the adjustment of the air supply speed and direction of the opened air inlet;
and adjusting the temperature and speed of the air supply based on the posture information and the activity intensity information of the indoor personnel.
8. The device for regulating and controlling personnel-oriented air supply based on image recognition as claimed in claim 5, wherein the regulation and control strategy aims at controlling pollutants meeting process cleanliness, and accordingly the air supply regulation and control module is specifically configured to:
adjusting the air inlet of the air supply based on the position information of the indoor personnel in the three-dimensional space, including opening the air inlet;
and adjusting the air supply speed based on the posture information and the activity intensity information of the indoor personnel.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method for regulating and controlling the supply of air to a person based on image recognition according to any one of claims 1 to 4 when executing the program.
10. A non-transitory computer-readable storage medium, on which a computer program is stored, wherein the computer program, when being executed by a processor, implements the steps of the method for regulating and controlling human supply air based on image recognition according to any one of claims 1 to 4.
CN202010707311.7A 2020-07-21 Personnel air supply-oriented regulation and control method and device based on image recognition Active CN111932630B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010707311.7A CN111932630B (en) 2020-07-21 Personnel air supply-oriented regulation and control method and device based on image recognition

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010707311.7A CN111932630B (en) 2020-07-21 Personnel air supply-oriented regulation and control method and device based on image recognition

Publications (2)

Publication Number Publication Date
CN111932630A true CN111932630A (en) 2020-11-13
CN111932630B CN111932630B (en) 2024-08-02

Family

ID=

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112907569A (en) * 2021-03-24 2021-06-04 北京房江湖科技有限公司 Head image area segmentation method and device, electronic equipment and storage medium
CN113865065A (en) * 2021-10-12 2021-12-31 珠海格力电器股份有限公司 Air outlet temperature control method and device, storage medium, processor and air supply equipment

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101504169A (en) * 2008-02-05 2009-08-12 Lg电子株式会社 Air conditioner and its control method
CN106989483A (en) * 2017-03-29 2017-07-28 邯郸美的制冷设备有限公司 Air blowing control method, system and the air conditioner of air conditioner
CN107421067A (en) * 2017-07-26 2017-12-01 广东美的制冷设备有限公司 Air-conditioning method, apparatus and computer-readable recording medium and air-conditioning
CN108537848A (en) * 2018-04-19 2018-09-14 北京工业大学 A kind of two-stage pose optimal estimating method rebuild towards indoor scene
CN109377513A (en) * 2018-09-20 2019-02-22 浙江大学 A kind of global credible estimation method of 3 D human body posture for two views
CN109987036A (en) * 2017-12-29 2019-07-09 周秦娜 A kind of control method improving interaction accuracy based on driver's body posture
CN111207499A (en) * 2020-01-09 2020-05-29 珠海格力电器股份有限公司 Air conditioner control method and air conditioner adopting same

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101504169A (en) * 2008-02-05 2009-08-12 Lg电子株式会社 Air conditioner and its control method
CN106989483A (en) * 2017-03-29 2017-07-28 邯郸美的制冷设备有限公司 Air blowing control method, system and the air conditioner of air conditioner
CN107421067A (en) * 2017-07-26 2017-12-01 广东美的制冷设备有限公司 Air-conditioning method, apparatus and computer-readable recording medium and air-conditioning
CN109987036A (en) * 2017-12-29 2019-07-09 周秦娜 A kind of control method improving interaction accuracy based on driver's body posture
CN108537848A (en) * 2018-04-19 2018-09-14 北京工业大学 A kind of two-stage pose optimal estimating method rebuild towards indoor scene
CN109377513A (en) * 2018-09-20 2019-02-22 浙江大学 A kind of global credible estimation method of 3 D human body posture for two views
CN111207499A (en) * 2020-01-09 2020-05-29 珠海格力电器股份有限公司 Air conditioner control method and air conditioner adopting same

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
WANG HUAN 等: "Airflow pattern induced by ceiling fan under different rotation speeds and blowing directions", 《INDOOR AND BUILT ENVIRONMENT》, 25 November 2019 (2019-11-25) *
张怡暄;徐士彪;孔彦;李红军;: "基于深度的虚实场景高清图像的融合方法", 中国体视学与图像分析, no. 03, 25 September 2013 (2013-09-25) *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112907569A (en) * 2021-03-24 2021-06-04 北京房江湖科技有限公司 Head image area segmentation method and device, electronic equipment and storage medium
CN112907569B (en) * 2021-03-24 2024-03-15 贝壳找房(北京)科技有限公司 Head image region segmentation method, device, electronic equipment and storage medium
CN113865065A (en) * 2021-10-12 2021-12-31 珠海格力电器股份有限公司 Air outlet temperature control method and device, storage medium, processor and air supply equipment

Similar Documents

Publication Publication Date Title
CN108764048A (en) Face critical point detection method and device
WO2020177498A1 (en) Non-intrusive human body thermal comfort detection method and system based on posture estimation
CN100360861C (en) Central air conditioner energy-saving control device based on omnibearing computer vision
CN102162667B (en) Image processing apparatus, image processing method, and air conditioning control apparatus
CN105972765B (en) A kind of intelligent air condition and its temperature control method
Tan et al. Digital Twin-driven approach to improving energy efficiency of indoor lighting based on computer vision and dynamic BIM
CN109887082A (en) A kind of interior architecture three-dimensional modeling method and device based on point cloud data
CN107655159A (en) A kind of air conditioner energy saving control method and system based on crowd density sensor model
CN107895080A (en) The big space air conditioning dynamic load estimation of fused images information and fresh air quantity control method
CN106960454A (en) Depth of field barrier-avoiding method, equipment and unmanned vehicle
CN110443776A (en) A kind of Registration of Measuring Data fusion method based on unmanned plane gondola
CN110377084A (en) A kind of Building Indoor Environment regulation method based on wisdom control strategy
CN111104904A (en) Human body identification method and device and courtyard machine
US20230083027A1 (en) Multi-area artificial fog pipe network intelligent control method and system based on yolov5 algorithm
WO2020135187A1 (en) Unmanned aerial vehicle recognition and positioning system and method based on rgb_d and deep convolutional network
Acquaah et al. Occupancy detection for smart HVAC efficiency in building energy: a deep learning neural network framework using thermal imagery
CN109145796A (en) A kind of identification of electric power piping lane fire source and fire point distance measuring method based on video image convergence analysis algorithm
CN110243381A (en) A kind of Lu Kong robot collaborative perception monitoring method
CN116300507B (en) Intelligent home management control system and method based on Internet of things technology
CN115272675A (en) Energy management system and method based on multi-sensor information fusion
CN114938556B (en) Automatic adjusting method and device for light of desk lamp, electronic equipment and storage medium
CN211427362U (en) Human body recognition device and courtyard machine
CN116797944A (en) Detection method and system for identifying cleanliness of photovoltaic panel based on unmanned aerial vehicle image
CN111832697A (en) Intelligent human body recognition energy-saving temperature regulating system
CN111932630B (en) Personnel air supply-oriented regulation and control method and device based on image recognition

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant