CN112969254B - Business hotel guest room illumination control device based on scene automatic identification - Google Patents

Business hotel guest room illumination control device based on scene automatic identification Download PDF

Info

Publication number
CN112969254B
CN112969254B CN202110233109.XA CN202110233109A CN112969254B CN 112969254 B CN112969254 B CN 112969254B CN 202110233109 A CN202110233109 A CN 202110233109A CN 112969254 B CN112969254 B CN 112969254B
Authority
CN
China
Prior art keywords
illumination
scene
value
lighting
guest room
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.)
Active
Application number
CN202110233109.XA
Other languages
Chinese (zh)
Other versions
CN112969254A (en
Inventor
黄昌清
邹细勇
井绪峰
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.)
Hefei Minglong Electronic Technology Co ltd
Original Assignee
China Jiliang 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 China Jiliang University filed Critical China Jiliang University
Priority to CN202110233109.XA priority Critical patent/CN112969254B/en
Publication of CN112969254A publication Critical patent/CN112969254A/en
Application granted granted Critical
Publication of CN112969254B publication Critical patent/CN112969254B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05B—ELECTRIC HEATING; ELECTRIC LIGHT SOURCES NOT OTHERWISE PROVIDED FOR; CIRCUIT ARRANGEMENTS FOR ELECTRIC LIGHT SOURCES, IN GENERAL
    • H05B45/00—Circuit arrangements for operating light-emitting diodes [LED]
    • H—ELECTRICITY
    • H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05B—ELECTRIC HEATING; ELECTRIC LIGHT SOURCES NOT OTHERWISE PROVIDED FOR; CIRCUIT ARRANGEMENTS FOR ELECTRIC LIGHT SOURCES, IN GENERAL
    • H05B45/00—Circuit arrangements for operating light-emitting diodes [LED]
    • H05B45/30—Driver circuits
    • H—ELECTRICITY
    • H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05B—ELECTRIC HEATING; ELECTRIC LIGHT SOURCES NOT OTHERWISE PROVIDED FOR; CIRCUIT ARRANGEMENTS FOR ELECTRIC LIGHT SOURCES, IN GENERAL
    • H05B45/00—Circuit arrangements for operating light-emitting diodes [LED]
    • H05B45/30—Driver circuits
    • H05B45/32—Pulse-control circuits
    • H05B45/325—Pulse-width modulation [PWM]
    • H—ELECTRICITY
    • H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05B—ELECTRIC HEATING; ELECTRIC LIGHT SOURCES NOT OTHERWISE PROVIDED FOR; CIRCUIT ARRANGEMENTS FOR ELECTRIC LIGHT SOURCES, IN GENERAL
    • H05B47/00—Circuit arrangements for operating light sources in general, i.e. where the type of light source is not relevant
    • H05B47/10—Controlling the light source
    • H—ELECTRICITY
    • H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05B—ELECTRIC HEATING; ELECTRIC LIGHT SOURCES NOT OTHERWISE PROVIDED FOR; CIRCUIT ARRANGEMENTS FOR ELECTRIC LIGHT SOURCES, IN GENERAL
    • H05B47/00—Circuit arrangements for operating light sources in general, i.e. where the type of light source is not relevant
    • H05B47/10—Controlling the light source
    • H05B47/105—Controlling the light source in response to determined parameters
    • H—ELECTRICITY
    • H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05B—ELECTRIC HEATING; ELECTRIC LIGHT SOURCES NOT OTHERWISE PROVIDED FOR; CIRCUIT ARRANGEMENTS FOR ELECTRIC LIGHT SOURCES, IN GENERAL
    • H05B47/00—Circuit arrangements for operating light sources in general, i.e. where the type of light source is not relevant
    • H05B47/10—Controlling the light source
    • H05B47/105—Controlling the light source in response to determined parameters
    • H05B47/115—Controlling the light source in response to determined parameters by determining the presence or movement of objects or living beings
    • H—ELECTRICITY
    • H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05B—ELECTRIC HEATING; ELECTRIC LIGHT SOURCES NOT OTHERWISE PROVIDED FOR; CIRCUIT ARRANGEMENTS FOR ELECTRIC LIGHT SOURCES, IN GENERAL
    • H05B47/00—Circuit arrangements for operating light sources in general, i.e. where the type of light source is not relevant
    • H05B47/10—Controlling the light source
    • H05B47/105—Controlling the light source in response to determined parameters
    • H05B47/115—Controlling the light source in response to determined parameters by determining the presence or movement of objects or living beings
    • H05B47/12—Controlling the light source in response to determined parameters by determining the presence or movement of objects or living beings by detecting audible sound
    • Y—GENERAL 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
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
    • Y02B20/00—Energy efficient lighting technologies, e.g. halogen lamps or gas discharge lamps
    • Y02B20/40—Control techniques providing energy savings, e.g. smart controller or presence detection

Landscapes

  • Circuit Arrangement For Electric Light Sources In General (AREA)

Abstract

The invention provides a business hotel guest room illumination control device based on scene automatic identification, which comprises an input module, an image acquisition module, a human body sensor, a processing module, an output module and a storage module, wherein the processing module further comprises an optimization processing unit, an event processing unit and a scene detection unit. The optimization processing unit is configured to establish an evaluation function for optimizing lamp parameters based on the illumination space model and the standard reaching level of illumination and uniformity in the guest room and the degree to which illumination requirements of various scenes such as offices and guests are met; the scene detection unit is configured to identify the illumination scene category based on the acquired image and the sensing data; the event processing unit is configured to output the optimized scene illumination control parameters corresponding to the event processing unit to the lamp according to the scene identification result, so that scene illumination based on automatic scene identification is realized, and automatic illumination with more pertinence, higher efficiency and energy conservation is obtained.

Description

Business hotel guest room illumination control device based on scene automatic identification
The application is a divisional application of application number 201811343930.1, application date 2018, 11 month 02, and invention name "business hotel guest room lighting control device based on scene automatic identification".
Technical Field
The invention belongs to the field of intelligent illumination, and particularly relates to a business hotel guest room illumination control device based on scene automatic identification.
Background
The hotel industry is a representative service industry and is concerned with the customer's in-house experience. Because the primary use area of customers is concentrated indoors and the use time is concentrated at night, the industry has a greater interest and urgent need for efficient, practical, personalized and automated intelligent lighting devices. The automatic scene switching function and the atmosphere rendering function of the intelligent lighting equipment are also interesting for the industry. In addition, in terms of energy consumption, the illumination energy consumption of the traditional business hotels often exceeds 30% of the total electricity consumption of the hotels, the energy consumption is relatively high, and after the intelligent illumination system is adopted, the intelligent illumination system can achieve a relatively effective energy-saving effect through modulating and integrally optimizing the brightness parameters of the single lamp, and the operation cost of the hotels is greatly reduced while the illumination requirements of users are met. Currently, over 70% of star commercial hotels are equipped with and use a variety of intelligent lighting devices in the hope that their customers will gain a more superior check-in experience.
However, the intelligent lighting devices used in the industry are often limited to simple infrared, voice-controlled switching devices, or wireless remote control devices, etc., and are generally conventional products with relatively wide application in the market. The product has wider application range, lacks a certain pertinence, lacks necessary researches on environment, layout and specific lighting requirements in the hotel industry, cannot completely meet and satisfy various special lighting requirements of business hotels, and cannot completely embody the self-superior lighting performance of the intelligent lighting system.
SUMMARY OF THE PATENT FOR INVENTION
The invention aims to provide a business hotel guest room illumination control device based on scene automatic identification, which improves illumination effectiveness and reduces unnecessary illumination on the basis of the illumination reference standard of the existing industry, thereby saving illumination power consumption. Specific lighting requirements under different scenes, such as offices, guests, entertainment and the like, are studied, different lighting requirements are used as guidance, optimization processing is adopted to optimize the light emitting distribution of the lamp, and optimized lighting control parameters are stored in the device. Meanwhile, a scene detection unit for training various lighting scenes by training the image set. Scene detection determines illumination scene identification according to the acquired image characteristics and the characteristic values of the sensing data, and based on the identified illumination scene, an output module sends a dimming signal to the lamp, and opens the corresponding scene to provide more targeted automatic scene illumination for a user.
The technical proposal of the invention is to provide a business hotel guest room lighting control device with the following structure, which comprises an input module, an image acquisition module, a human body sensor, a processing module, an output module and a storage module,
the processing module comprises an optimization processing unit, an event processing unit and a scene detection unit, wherein the optimization processing unit is configured to:
Responding to an optimized processing instruction of an event processing unit, based on an illumination space model which is formed by describing geometric parameters of each area in a business hotel guest room and light distribution parameters of each lamp distributed in the guest room, scoring the standard reaching degree of illumination and illumination uniformity of each area in the business hotel guest room by taking the illumination standard or general requirement of the business hotel industry as a reference, establishing an evaluation function F of an illumination effect, and supplementing the evaluation function F according to the scene use requirement of the business hotel illumination, thereby establishing an overall evaluation function F; optimizing the illumination control parameters of each lamp in the guest room by adopting particle swarm optimization based on the overall evaluation function F, and mapping the optimization result into scene illumination control parameters;
the scene detection unit is configured to:
training based on a training image set, extracting image characteristics and characteristic values of sensing data based on the image acquired by an image acquisition module and the human body sensor sensing data corresponding to the image, determining illumination scene identification of the image according to the image characteristics and the characteristic values of the sensing data, and transmitting an identification result to an event processing unit;
the event processing unit is configured to:
And responding to the signal input of the input module, displaying the processed signal to a user through the output module, and responding to the identification result signal of the scene detection unit, and outputting the scene lighting control parameter corresponding to the identification result acquired by the optimization processing unit.
Preferably, the system further comprises an illuminance acquisition module, wherein the illuminance acquisition module acquires illuminance information from a plurality of illuminance observation points and transmits the illuminance information to the processing module.
Preferably, the output module comprises a communication interface unit, which is connected to the driver of the luminaire.
Preferably, the input module further comprises a control panel, and the event processing unit is further used for responding to the operation of opening the lighting scene on the control panel by a user, converting the lighting control parameters corresponding to the scene into driving current values of the lamp, and transmitting the driving current values to a driver of the lamp through a communication interface unit in the output module.
Preferably, the system further comprises a storage module, wherein the storage module stores the illumination space model of the business hotel guest room, the particle swarm optimization processing parameters and the scene illumination control parameters optimized by the processing module.
Preferably, the scene detection unit includes a feature extraction section, a recognition section, and a training section,
the feature extraction section extracts an image and a sensor feature,
the training part trains the untrained recognition part according to the extracted characteristics and the scene category corresponding to the characteristics based on the training image set, thereby obtaining the recognition part,
the recognition part recognizes the illumination scene according to the extracted features, thereby obtaining illumination scene recognition of the image,
the storage module also stores structural parameters of the scene detection unit.
The optimization of the processing module is based on the scoring of the light distribution in the hotel room, and the evaluation function on which the scoring is based depends on the establishment of the illumination space model. Preferably, the optimization processing unit is further configured to:
dividing the business hotel guest room into three areas, namely a working area, a nearby area and a background area, according to the use frequency of the business hotel guest room, wherein the areas with more uses, such as a bed surface, a writing desk, a sofa and the like, are defined as working areas; areas that other users may use, but do not stay for a long time, are listed as areas adjacent to the work plane; the area between the bed surface and the wall of the toilet and the position of the windowsill are listed in the background area;
Taking the illuminance and the uniformity of the illuminance of each area as main indexes, and establishing an evaluation function f in an optimization processing unit:
f＝w 1 ×u(E 1 )+w 2 ×u(E 2 )+w 3 ×u(E 3 )+w 4 ×u(U 1 )+w 5 ×u(U 2 )
wherein E is 1 For the horizontal illumination of the working plane E 2 For illumination of the vicinity of the working plane, E 3 For background area illumination, U 1 For uniformity of illumination of working plane, U 2 Illuminance uniformity of a nearby area of the working plane; w (w) 1 、w 2 、w 3 、w 4 、w 5 Respectively assigning weights of the indexes by using an analytic hierarchy process and inputting the weights through an input module; u () represents the degree of proximity between the calculated value obtained from the luminaire light distribution model and the reference value for each index.
The analytic hierarchy process is as follows:
establishing a judgment matrix A according to the number h of the weight coefficients h*h And uses interval [1,9 ]]Filling the comparison matrix of the two integers, wherein the numerical value of each element represents the importance degree of the index corresponding to the corresponding row weight item compared with the index corresponding to the corresponding column weight item, and the numerical value can be a subjective evaluation result or a statistical result obtained through sample investigation; then, matrix A h*h Substituting into a geometric mean equation, determining each weight coefficient value, wherein the geometric mean equation is as follows:
the functional area division of the business hotel guest room is not included in the washroom, because the influence of the washroom on other areas is small, and the washroom has little influence on the outside under the condition that the door of the washroom is closed, so the washroom is not included in the research scope. The sub-function u () in the evaluation function f represents each under a specific lighting control parameter configuration The degree of closeness of the calculated value of the term index to the reference value obtained by the relevant standard or suggestion, for E 1 、E 2 、E 3 In other words, u () can be expressed by the following expression when the calculated value is not equal to the reference value:
wherein E is a reference value of the corresponding evaluated index, and E' is an illuminance calculation value of each area when the lighting lamp performs lighting according to the corresponding lighting control parameters. For U 1 、U 2 In other words, when the calculated value is smaller than the reference value, the calculation is performed according to the first ratio formula, otherwise, when the calculated value is larger than the reference value, u () =1 is still considered.
The evaluation function f evaluates the standard reaching degree of illuminance uniformity of each area. Preferably, the optimization processing unit is further configured to:
aiming at different illumination of business hotels and specific use requirements, such as energy conservation, office, guest reception, entertainment, leisure reading and the like, the method respectively supplements F to establish a new evaluation function F 1 To F 5 To enhance the pertinence of the lighting effect and reduce the energy consumption of the lamp to a certain extent:
firstly, taking lamp illumination energy conservation as an optimization target, supplementing an evaluation function F, wherein the lamp energy consumption is reduced to the minimum while the illumination and illumination uniformity indexes are ensured in each working area, and if n lamps are shared in a guest room, a new evaluation function F is provided 1 Can be expressed as:
wherein P (i) is the power consumption of the ith lamp;
secondly, aiming at office scenes, the illumination of a writing desk and illumination uniformity are required to meet reference conditions as far as possible, and a new evaluation function F is obtained 2 Can be expressed as:
wherein E is 11 Representing the horizontal illumination of the desktop in the writing desk area, U 11 Representing the uniformity of the illumination of a desktop in a writing desk area, q 1 、q 2 、k 1 、k 2 、k 3 The weight coefficient of the corresponding item can be obtained by a hierarchical analysis method;
thirdly, aiming at the guest receiving scene, the overall brightness and illumination uniformity in the required area are more similar, the illumination reference value in each area is uniformly set to be a fixed value, the illumination uniformity reference value is set to be a fixed value, and a new evaluation function F is provided 3 Expressed as:
wherein k is 1 、k 2 The weight coefficient of the corresponding item can be obtained by a hierarchical analysis method;
fourth, for entertainment scenes, a lower illumination intensity of the background and the nearby area is required, and a new evaluation function F is obtained 4 Expressed as:
wherein E is 12 Represents the horizontal illumination of the writing desk outside the working area, q 1 、q 2 、q 3 、k 1 、k 2 、k 3 The weight coefficient of the corresponding item can be obtained by a hierarchical analysis method;
fifthly, aiming at leisure reading scenes, the brightness of m lamps near the head of the bed is required to be higher, and a new evaluation function F is adopted 5 Expressed as:
wherein I is t Respectively representing the brightness of the t lamp near the head of the bed,q t representing their respective weight coefficients; k (k) 1 、k 2 、k 3 The values of the weight coefficients for the respective terms can be obtained by a hierarchical analysis method.
The optimizing unit is based on each evaluation function F i And (i=1, 2.) 5, optimizing the lighting control parameters of each lamp in the guest room by adopting particle swarm optimization, and respectively converting a plurality of optimization results obtained according to different evaluation functions into the control parameters of the lighting scene of the lamp to store so as to be used for automatic scene lighting control after scene identification.
The particle swarm optimization processing flow is as follows:
s1, randomly generating an initial set X (1) of particle groups N*n Its element x ij (1) I=1, 2,..n, j=1, 2,..n; n is the number of lamps to be optimized in the guest room, namely the dimension of the particle swarm, and N is the size of the particle swarm; each element x ij (1) The i-group particle initial illumination control parameter modulation value of the lamp j in the guest room is represented, the illumination distribution of each region of the guest room in the hotel is calculated for all n lamps corresponding to the i-group particle according to the light distribution model when each lamp illuminates according to the current illumination control parameter modulation value, and an evaluation function F is adopted 1 Evaluating the illumination effect corresponding to the ith group of particles;
respectively obtaining initial grading values of the lighting control parameter modulation value vectors of each particle group, taking the initial grading values as initial grading values of self-history optimal solutions, and recording the lighting control parameter modulation value vectors;
meanwhile, the maximum value in all N groups of particle initial scoring values is recorded as the initial scoring value of the group history optimal solution, and the illumination control parameter modulation value vector is recorded, so that the updating frequency k=1;
s2, in the particle swarm optimization process, each element x ij () All corresponding to a change value v ij () Element v ij () Modulation value x of illumination control parameter of corresponding lamp j ij () V is also needed to be carried out on the basis of (1) ij () To ensure its effectiveness, v ij () The value interval should not be larger than x ij () 10% of the maximum value that can be taken; if change toChanged x ij () Greater than x ij () Maximum value possible to be taken, x ij () Still taking its maximum value;
randomly generated set V (k) N*n The elements thereof respectively represent a collection X (k) N*n The values of the elements are changed according to the following formula N*n X (k) N*n Performing multiple updates:
v ij (k+1)＝wv ij (k)+c 1 r 1 (P ij (k)-x ij (k))+c 2 r 2 (G j (k)-x ij (k))
x ij (k+1)＝x ij (k)+v ij (k+1)
wherein k is the current update times; p (P) ij (k) The specific illumination control parameter modulation values corresponding to the self-history optimal solution of the ith group of particle data in the kth updating are represented, the specific illumination control parameter modulation values are obtained by comparing the score values of the updated illumination control parameter modulation values and the self-history optimal solution, and if the updated score values are larger than the score values of the self-history optimal solution, the updated illumination control parameter modulation values are recorded as new self-history optimal solutions;
G j (k) The particle group history optimal solution representing the jth specific illumination control parameter modulation value in the kth updating is obtained by comparing the updated illumination control parameter modulation values of each group with the scoring value of the group history optimal solution;
c 1 ，c 2 is constant, r 1 ，r 2 Is a random number uniformly distributed between the intervals (0, 1); w is a weight coefficient; w (w) max ，w min Respectively a maximum value and a minimum value of weight coefficients, X (k+1) N*n V (k+1) N*n Representation pair X (k) N*n V (k) N*n Is the kth optimization update of (c);
k represents the maximum number of updates, and when K equals K, the update is ended and G is output j (k) In (a)And the recorded illumination control parameter modulation value of the group history optimal solution is used as a final output result.
S3, repeating the optimization process described in S1 and S2 four times, and respectively utilizing the evaluation function F 2 To F 5 Replacing the evaluation function F in S1 one by one 1 Five groups of different optimization results are obtained, and the optimization results are converted and stored into lighting scene lamp regulation parameters for users to use.
And (3) acquiring sample images of a user in various illumination scenes such as office, reading, receiving, entertainment and energy conservation while optimizing illumination control parameters in the illumination scenes, and forming a training image set by the sample images to train the scene detection unit.
After training is completed, an image of a business hotel guest room and sensing data corresponding to the image are acquired, a scene detection unit extracts image features and feature values of the sensing data, and illumination scene identification of the image is determined according to the image features and the feature values of the sensing data.
Preferably, the sample image may also leave a portion as a verification set to verify the trained scene detection unit.
Preferably, the scene detection unit may employ a support vector machine of a linear kernel function, i.e., an SVM classifier.
Preferably, according to the characteristic values of the collected image characteristics and the sensing data, the lighting scene identification of the image is determined specifically as follows: determining one or more lighting scene identifications of the image according to the image features and feature values of the sensing data; when the determined illumination scene of the image is identified as a plurality, the illumination scene of the image is determined to be identified as a composite illumination scene.
Preferably, the lamp adjustment parameter may be a lamp luminous flux or a driving current corresponding to the lamp luminous flux, or may be a driving current corresponding to other parameters such as a lamp color temperature.
The magnitude of the drive current may be an absolute value or a percentage value with respect to the rated drive current value.
Parameters such as group size, adjustment coefficient, evaluation function and the like involved in the optimization processing of the particle swarm, and scene illumination control parameters, structural parameters of a scene detection unit and the like optimized by the processing module can be respectively input through a keyboard in the input module or extracted after the optimization or training is finished and stored in the storage module.
Preferably, the input/output module may be integrated on a main body, such as a touch screen of a communication interface. The lamps in the hotel room adopt the dimmable lamps, such as the dimmable LED lamp group, and the optimized illumination control parameters under various scene applications are converted into signals for adjusting the light output of the lamp group, such as the driving current of each LED lamp in the lamp group. Typically, the light emission brightness of the passenger room can be adjusted by adjusting the PWM duty ratio of the driving current, so that the illumination of the target illumination surface in the passenger room is changed.
Preferably, an automatic scene control button is arranged on a control panel of the output module, and the button is pressed to automatically identify the scene and the corresponding illumination scene is opened according to the optimized illumination control parameters by the post-processing module.
Preferably, the event processing unit is further configured to: and responding to a signal of automatic identification completion of the lighting scene, converting the optimized lighting control parameter corresponding to the identified scene into a PWM wave duty ratio value of the lamp driving current, and transmitting the PWM wave duty ratio value to a driver of the lamp through an output module.
Preferably, when the lighting scene is identified as the integrated lighting scene identification, the adjustment control of the power supply current of the LED lamp group is performed according to the weighted sum of the lamp adjustment parameters corresponding to the plurality of lighting scenes.
Preferably, before the illumination scene identification is completed, the processing module may illuminate the guest room according to a preset illumination scene, for example, all lights of the guest room may be turned on or turned on to 80% brightness.
Preferably, the control panel in the input module can also receive manual adjustment operation, and the power supply current of the LED lamp group is manually adjusted through the processing module and the output module.
The invention has the advantages that:
1. the lighting control parameters are designed according to the lighting requirements of the business hotels, and more targeted automatic scene lighting is provided for users;
2. a plurality of different target scene optimization indexes are provided to meet different use demands of users, so that the method is more humanized;
3. the optimization index meets the illumination requirement of a user, and simultaneously gives consideration to the energy consumption index, so that the energy is saved and the environment is protected;
4. the operation is convenient, and the automatic switching control and the manual adjustment can be realized in various different scenes.
Drawings
FIG. 1 is a block diagram of the structure of the present invention;
FIG. 2 is a schematic diagram of a scene detection unit according to the present invention;
FIG. 3 is a schematic view of a guest room area according to the present invention;
FIG. 4 is a schematic diagram of the distribution of the LED lamps and illuminance observation points according to the present invention;
FIG. 5 is a schematic diagram of a lamp distribution according to the present invention;
FIG. 6 is a graph showing score updating in the optimization process of the present invention;
fig. 7 is a schematic distribution diagram of a camera and a control panel according to the present invention.
Wherein, the bed surface 1, the writing desk 2, the sofa 3, the adjacent area 4, the background area 5, the toilet 6, the hand-washing desk 7, the hall 8, the windowsill 9 and the camera 10,
the device comprises a 100 lighting control device, a 110 input module, a 111 keyboard, a 112 control panel, a 120 illumination acquisition module, a 130 processing module, a 131 optimization processing unit, a 132 event processing unit, a 133 scene detection unit, a 1331 feature extraction unit, a 1332 recognition unit, a 1333 training unit, a 140 storage module, a 141 lighting space model, a 142 scene lighting control parameter set, a 143 scene detection unit structure parameter set, a 150 output module, a 151 display screen, a 152 communication interface unit, a 160 image acquisition module, a 170 human body sensor, a 200 driver, a 300 LED lamp and a 400 illumination observation point.
Detailed Description
The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, but the present invention is not limited to these embodiments only. The invention is intended to cover any alternatives, modifications, equivalents, and variations that fall within the spirit and scope of the invention.
In the following description of preferred embodiments of the invention, specific details are set forth in order to provide a thorough understanding of the invention, and the invention will be fully understood to those skilled in the art without such details.
The invention is more particularly described by way of example in the following paragraphs with reference to the drawings. It should be noted that the drawings are in a simplified form and are not to scale precisely, but rather are merely intended to facilitate and clearly illustrate the embodiments of the present invention.
As shown in fig. 1, the lighting control device 100 for a business hotel guest room based on scene automatic identification of the present invention includes an input module 110, an image acquisition module 160, a human body sensor 170, a processing module 130, and an output module 150. Wherein the input module 110 receives parameter input and user operation through a keyboard 111 and a control panel 112 included therein, such as parameters including: geometric positions and/or shapes of functional areas of a business hotel guest room in the illumination space model in a world coordinate system, geometric positions of illumination lamps and light distribution models of the lamps; in addition, parameters such as group size, adjustment coefficient, iteration number, weight in an evaluation function and the like which are involved in the particle swarm optimization process are also included, and scene illumination control parameters, lamp driving current, structural parameters of a scene detection unit and the like which are optimized by the processing module are also included. In addition to other parameters required for the operation of the device, such as communication interface parameters, etc. User operations, including clicking, touching, etc., on the control panel.
Meanwhile, the parameters are stored in the storage module 140, wherein the parameters of the illumination space model are stored in the illumination space model 141 in a custom data structure, the optimized illumination control parameters of each lamp, such as the light flux modulation value, are also stored in the scene illumination control parameter set 142 in a custom data structure, and the structural parameters of the scene detection unit after training and the rule base for identifying the illumination scene of the image are stored in the scene detection unit structural parameter set 143.
The body sensors 170 are positioned in areas of the guest room where the user passes or stays, such as doorways and locations near desks, sofas, and bed heads. The human body sensor 170 provides a sensing signal of scene detection to the scene detection unit 133 together with the image acquisition module 160.
The processing module 130 includes an optimizing processing unit 131, an event processing unit 132 and a scene detecting unit 133, wherein the optimizing processing unit 131 evaluates and optimizes various light distributions based on reference values of illuminance and illuminance uniformity of different target areas on the basis of performing area division on a business hotel guest room and requirement definition on different application scene illumination distributions, and converts an optimizing result into illumination control parameters of each lamp in the guest room, and specifically, the optimizing processing unit 131 is configured to:
Responding to the optimized processing instruction of the event processing unit 132, based on an illumination space model which describes geometric parameters of each area in a business hotel guest room and light distribution parameters of each lamp distributed in the guest room, scoring the standard reaching degree of illumination and illumination uniformity of each area in the business hotel guest room by taking the illumination standard or general requirement of the business hotel industry as a reference, establishing an evaluation function F of illumination effect, and supplementing the evaluation function F according to the scene use requirement of the business hotel illumination, thereby establishing an overall evaluation function F; and optimizing the illumination control parameters of each lamp in the guest room by adopting particle swarm optimization based on the overall evaluation function F, and mapping the optimization result into scene illumination control parameters.
The scene detection unit 133 is configured to: the training image set is trained based on the training image set, the image features and the feature values of the sensing data are extracted based on the image acquired by the image acquisition module 160 and the sensing data from the human body sensor 170 corresponding to the image, the illumination scene identification of the image is determined according to the image features and the feature values of the sensing data, and the identification result is sent to the event processing unit 132.
The event processing unit 132 is configured to: the processed signal is displayed to the user through the output module 150 in response to the signal input of the input module 110, and the scene lighting control parameter corresponding to the recognition result acquired through the optimization processing unit 131 is output in response to the recognition result signal of the scene detection unit 133. The method specifically comprises the following steps: when a user inputs parameters through the keyboard 111 in the input module 110, the operation interaction information is displayed through the display screen 151, and the parameters are transmitted to the storage module 140 to be stored; the event processing unit 132, after receiving an optimization instruction preset after the user issues or starts through the input module 110, notifies the optimization processing unit 131 to perform the optimization process, and the optimization result is saved. Meanwhile, after the lighting scene identification of the scene detection unit 133 is completed, the event processing unit 132 responds, and transmits the lighting control parameters corresponding to the optimization result of the scene corresponding to the identification result to the lamps through the output module 150 in the form of instructions or message notification, so that the lamps in the guest room perform light output adjustment according to the optimization result, thereby realizing optimized lighting in various scenes and improving user experience.
As shown in fig. 2, the scene detection unit 133 includes a feature extraction section 1331, a recognition section 1332, and a training section 1333. The feature extraction unit 1331 extracts the image and the sensor feature, and the training unit 1333 trains the recognition unit 1332 based on the training image set based on the extracted feature and the scene type corresponding to the feature, thereby obtaining the recognition unit 1332 capable of recognizing the illumination scene of the image. After training, the recognition unit 1332 recognizes the illumination scene based on the features extracted by the feature extraction unit 1331 with respect to the acquired image and sensing data of the illumination scene in the current guest room, thereby obtaining the current illumination scene recognition.
As shown in fig. 3 and 7, to illustrate the application of the lighting control device for the business hotel guest room of the invention in optimizing control of lighting, a hotel guest room with a length and width of 6m×4m×3m is selected as an example. Dividing a business hotel guest room into a plurality of areas according to use requirements, wherein the bed surface 1, the writing desk 2 and the sofa 3 are areas which are used by users more, so that the business hotel guest room is defined as a working area; other parts, areas that the user may use, but do not stay for a long time, are listed as adjacent areas 4; the area between the bed surface 1 and the wall of the toilet 6 and the position of the windowsill 9 are listed in the background area 5 because the user generally uses little or has low illumination requirements when using the device; the washroom 6 includes the washstand 7, and is not included in the study since it has little influence on other areas, and hardly has any influence on the outside when the door of the washroom is closed.
As shown in fig. 3 and 4, eight dimmable LED lamps 300 marked with cross circles are placed in the guest room in total, and sixteen illuminance observation points 400 marked with cross rectangles are set for observing illuminance distribution in each functional area.
After the regional division is finished, the lighting environment of the business hotel is mathematically modeled by consulting the relevant standard or advice of the lighting of the hotel, and the geometric positions and/or shapes of each functional region of the guest room and each lighting lamp are represented in a world coordinate system; meanwhile, acquiring a light distribution model of each lamp; then, an evaluation function f is established, and the standard reaching degree of the calculated value of the illuminance and the illuminance uniformity of each area relative to the reference value is scored:
f＝w 1 ×u(E 1 )+w 2 ×u(E 2 )+w 3 ×u(E 3 )+w 4 ×u(U 1 )+w 5 ×u(U 2 )
wherein E is 1 Is the horizontal illumination (unit: lx) of the working plane, and the general reference value is 300, E 2 For illumination of the vicinity of the working plane, the general reference value is 260, E 3 For background area illumination, its general reference value is 200U 1 For uniformity of illumination of the working plane, the general reference value is 0.7, U 2 Taking a general reference value of 0.6 for the illuminance uniformity of the adjacent area of the target working plane; w (w) 1 、w 2 、w 3 、w 4 、w 5 Respectively the weight coefficients of the indexes; u () represents the proximity of the calculated value of each index to the reference value, for E 1 、E 2 、E 3 In other words, u () can be expressed by the following expression when the calculated value is not equal to the reference value:
wherein E is a reference value of the corresponding evaluated index, and E' is an illuminance calculation value of each area when the lighting lamp performs lighting according to the corresponding lighting control parameters. For U 1 、U 2 In other words, when the calculated value is smaller than the reference value, the calculation is performed according to the first ratio formula, otherwise, when the calculated value is larger than the reference value, u () =1 is still considered.
The weight coefficient w in the evaluation function f is subjected to analytic hierarchy process 1 、w 2 、w 3 、w 4 、w 5 Assigning, firstly establishing a judgment matrix A according to the number h of weight coefficients h*h Filling the matrix with the ratio of the two integers from 1 to 9, wherein the value of the matrix represents the index corresponding to the corresponding row weight item, and compared with the importance degree (1 is the same importance as the two indexes, a) of the index corresponding to the corresponding column weight item ij For the importance ratio of the ith weight term to the jth weight term, e.g. a ij A value less than 1 indicates that the latter j is more important than the former i). Without loss of generality, matrix A h*h The numerical values of the elements are set by subjective evaluation, and the numerical values are as follows:
wherein, rows 1 to 5 and columns 1 to 5 respectively correspond to the optimization index E 1 、E 2 、E 3 、U 1 、U 2 . Element a 12 A value of 5, i.e. indicating that the filling decision considers E 1 Is of higher importance than E 2 The method comprises the steps of carrying out a first treatment on the surface of the At the same time, there is a certain independence between the elements, e.g. E 2 And E is connected with 3 The importance relationship between the elements should not be defined by element a 12 And a 13 Is determined by the ratio of (2). Then, matrix A 5*5 Substituting the specific numerical value of each weight coefficient into a geometric mean equation, wherein the geometric mean equation is as follows:
the index weight coefficients can be obtained as follows:
the analytic hierarchy process can be used for calculating the weight coefficient of the model, and the model can be input and stored in a storage module after offline completion, or the optimization processing unit can be used for completing the calculation according to the input judgment matrix.
The desired illuminance distribution for the occupant varies under different lighting scene requirements. Therefore, on the basis of the grading based on the standard reaching degree, the grading calculation formula needs to be correspondingly adjusted so as to reflect the parameter index optimization requirement corresponding to the specific scene. According to different use demands of guest rooms, five scenes such as energy conservation, guest reception, entertainment, office, leisure reading and the like are determined as optimization target scenes, so that the living experience of a living person is improved. For different scenes, respectively supplementing F according to scene demand characteristics to establish a new evaluation function F 1 To F 5 At this time, the reference values cited in the indexes can be adjusted on the basis of the general reference values, and the weight coefficients of the indexes can be reassigned according to the preference.
Firstly, for the energy-saving index, the overall energy consumption of the lamp is required to be reduced as much as possible, so that the corresponding evaluation function F is adopted 1 Expressed as:
wherein P (i) is the power consumption of the ith lamp, it can be seen that the smaller P (i), the evaluation function F 1 The higher the value of (c).
Secondly, aiming at office scenes, for users needing to office, the office area is mainly concentrated in the desk area, and in order to achieve the best working illumination effect, the illumination of the desk and the illumination uniformity are required to meet the reference condition as far as possibleNew evaluation function F at this time 2 Can be expressed as:
wherein E is 11 Representing the horizontal illumination of the desktop in the writing desk area, U 11 Representing the uniformity of the illumination of a desktop in a writing desk area, q 1 、q 2 、k 1 、k 2 、k 3 The weight coefficient of the corresponding item can be obtained by a hierarchical analysis method.
Wherein E is 11 The table top horizontal illumination in the writing table area is represented, and the reference value is 320; u (U) 11 The uniformity of the illumination of the desktop in the writing desk area is represented, and the reference value is taken as 0.7; q 1 、q 2 、k 1 、k 2 、k 3 The weight coefficients of the corresponding terms can also be obtained by a hierarchical analysis method, and the values can be 0.45, 0.55, 0.27, 0.46 and 0.27 respectively without losing generality.
Thirdly, regarding the evaluation index, the overall brightness and illuminance uniformity in the region are more focused on the scene, thus uniformly setting the illuminance reference value in each region to be a fixed value such as 350lx and the illuminance uniformity reference value to be a fixed value such as 0.6 to create a bright and comfortable lighting atmosphere, and the new evaluation function F 3 Expressed as:
by increasing the weighting coefficient k 1 Is taken to be k 1 ＞k 2 To increase the overall illuminance and illuminance uniformity requirements during the evaluation process. Re-aligning w by analytic hierarchy process 1 、w 2 、w 3 、w 4 、w 5 、k 1 、k 2 Assignment was made to values 0.2765, 0.2559, 0.0999, 0.2765, 0.0922, 0.71, 0.29, respectively.
Fourth, amusement scene, the scene is mainly suitable for the user to be located in bedWhen the mobile phone or the television is used, the illumination brightness of the background and the adjacent area is required to be low, the illumination reference value can be set to 150lx, and the new evaluation function F 4 Expressed as:
wherein E is 12 Represents the horizontal illumination of the writing desk outside the working area, q 1 、q 2 、q 3 、k 1 、k 2 、k 3 Is the weight coefficient of the corresponding item. According to the analytic hierarchy process, w 1 、w 2 、w 3 、w 4 、w 5 、q 1 、q 2 、q 3 、k 1 、k 2 、k 3 Assigned as 0.2765, 0.2559, 0.2765, 0.0999, 0.0922, 0.27, 0.39, 0.34, 0.37, 0.41, 0.12, respectively.
Fifthly, the leisure reading scene is mainly used for a user to read books in a bedside or sofa area, and requires that m lamps near the bedside have higher brightness, and a new evaluation function F is obtained at the moment 5 Expressed as:
wherein I is t Respectively represents the brightness, q of the t-th lamp near the head of the bed t Representing their respective weight coefficients; k (k) 1 、k 2 、k 3 Is the weight coefficient of the corresponding item.
In combination with the figures 3 and 4, the leisure reading scene requires that the brightness of f, g and h lamps near the head of the bed is higher. According to the analytic hierarchy process, w 1 、w 2 、w 3 、w 4 、w 5 、q 1 、q 2 、q 3 、k 1 、k 2 、k 3 0.1338, 0.0933, 0.0780, 0.3662, 0.3286, 0.33, 0.37, 0.41, 0.12, respectively.
Evaluation function F 1 To F 5 Establishment ofAfter the optimization, the particle swarm optimization is adopted to optimize the parameters of the lamp, namely the illumination control parameters or the modulation values thereof, and the evaluation functions F are respectively used in the optimization process 1 To F 5 Each particle in the population was evaluated.
The particle swarm optimization is used as an effective intelligent search algorithm, is widely applied to searching applications of optimal values of various discrete points, and has the advantages of strong optimizing capability, higher flexibility of an implementation method and wide application field. The basic principle can be regarded as a group of particles in the search space, and the positions of the particles are possible values of the optimal values of the optimized indexes. By tracking the change of the particle position, the optimal solution (Pbest) and the group optimal solution (Gbest) of the particle are searched, and the position of the particle is adjusted according to the two indexes.
As shown in fig. 4 and 5, eight dimmable LED lamps 300, namely lamps a to h, are placed in a guest room, a certain brand of LED bulbs are used, the rated power is 15W, the theoretical illuminance is 1600lm, the layout positions of the lamp bulbs are placed according to the original lamp holder positions of a commercial hotel, sixteen illuminance observation points 400 are set, and the illuminance observation points 400 are ensured to be positioned at the vertical center line or the horizontal center line of each region as much as possible and are distributed at equal intervals during selection. Illuminance observation points can be added as needed.
The light distribution model of the luminaire is preferably validated, and if there is no prior model, it can be obtained by using an actual individual lighting experiment.
Referring to fig. 1, preferably, an illuminance acquisition module 120 is further provided, which acquires illuminance information from a plurality of illuminance observation points and transmits the illuminance information to the processing module. The event processing unit responds to the acquisition input of the illumination information, calculates and stores the light emitting data of each lamp, and establishes a light distribution model of the lamp. Then, according to the light distribution model of the lamp, the relative positions of the lamp and the sampling points in the world coordinate system, calculating: the eight dimmable LED lamps 300 modulate the illumination at each sampling point in space with a specific lamp parameter, i.e., the illumination control parameter. The sampling points are arranged in each divided area to obtain illuminance and illuminance uniformity of each area, and then the evaluation value of the illumination control parameter modulation value can be calculated.
Based on an evaluation function F 1 To F 5 The flow of the particle swarm optimization processing adopted by the optimization processing unit is as follows:
s1, randomly generating a set X (1) 50*8 Which represents eight lamps in a guest room, the particle swarm is 50 groups in size, each element x ij (1) I groups of particle initial illumination control parameter modulation values of a lamp j in a guest room are represented;
calculating the illuminance distribution of each area of the hotel guest room when each lamp is illuminated according to the current illumination control parameter modulation value according to all 8 lamps corresponding to the ith group of particles according to the light distribution model, and evaluating the illumination effect corresponding to the ith group of particles by adopting a current evaluation function F;
the illumination of each area takes the value according to the average illumination value of all sampling points in the area, and the illumination uniformity is calculated according to the illumination of the sampling points:
illuminance uniformity = minimum illuminance value/average illuminance value
The corresponding relation of the illumination values of the sampling points is as follows:
E jP for the horizontal illumination of any point P in the illumination space of the lamp j, L and H are the distances between the point P and the lamp in the horizontal and vertical directions respectively; i θ The light intensity in the theta direction when the total luminous flux of the lamp is 1000lm is represented, and K is a maintenance coefficient; phi is the actual luminous flux of the lamp, namely, the set X (1) 50*8 Each element x of ij (1) Corresponding illumination control parameter modulation values;
respectively obtaining initial grading values of the lighting control parameter modulation value vectors of each particle group according to the evaluation function, taking the initial grading values as initial grading values of self-history optimal solutions, and recording the lighting control parameter modulation value vectors;
Meanwhile, the maximum value in all N groups of particle initial scoring values is recorded as the initial scoring value of the group history optimal solution, and the illumination control parameter modulation value vector is recorded, so that the updating frequency k=1;
s2, in the particle swarm optimization process, each element x ij () All corresponding to a change value v ij () Element v ij () Modulation value x of illumination control parameter of corresponding lamp j ij () V is also needed to be carried out on the basis of (1) ij () To ensure its effectiveness, v ij () The value interval should not be larger than x ij () 10% of the maximum value that can be taken; if x is changed ij () Greater than x ij () Maximum value possible to be taken, x ij () Still taking its maximum value;
randomly generated set V (k) 50*8 The elements thereof respectively represent a collection X (k) 50*8 The values of the elements are changed according to the following formula 50*8 X (k) 50*8 Performing multiple updates:
v ij (k+1)＝wv ij (k)+c 1 r 1 (P ij (k)-x ij (k))+c 2 r 2 (G j (k)-x ij (k))
x ij (k+1)＝x ij (k)+v ij (k+1)
wherein k is the current update times; p (P) ij (k) The specific illumination control parameter modulation values corresponding to the self-history optimal solution of the ith group of particle data in the kth updating are represented, the specific illumination control parameter modulation values are obtained by comparing the score values of the updated illumination control parameter modulation values and the self-history optimal solution, and if the updated score values are larger than the score values of the self-history optimal solution, the updated illumination control parameter modulation values are recorded as new self-history optimal solutions;
G j (k) The particle group history optimal solution representing the jth specific lighting control parameter modulation value in the kth updating is obtained by comparing the updated lighting control parameter modulation values of each group with the scoring value of the group history optimal solution, and if the updated scoring value of a certain group is larger than the scoring value of the group history optimal solution, the particle group history optimal solution is obtained by comparing the updated lighting control parameter modulation value of each group with the scoring value of the group history optimal solutionThe updated illumination control parameter modulation values are recorded as new group history optimal solutions; the method comprises the steps of carrying out a first treatment on the surface of the
c 1 ，c 2 The constant is generally 2.05, r 1 ，r 2 For random numbers uniformly distributed between the intervals (0, 1), it can be obtained by a random number generation function; w is a weight coefficient, w max ，w min The maximum value and the minimum value of the weight coefficient are respectively taken as 0.9 and 0.5.
X(k+1) i*n V (k+1) i*n Representation pair X (1) i*n V (1) i*n The k-th optimization result of (c). K represents the maximum number of updates, and when K equals K, the update is ended and G is output i (k) The illumination control parameter modulation value of the group history optimal solution recorded in the above is used as the final output result.
The optimization process described in S1 and S2 is repeated four times, wherein in S1, 1 is replaced by k and the initial random generation process is skipped, and then the evaluation function F is utilized respectively 2 To F 5 Replacing the evaluation function F in S1 one by one 1 Five groups of different optimization results are obtained, and the optimization results are converted and stored into lighting scene lamp regulation parameters for users to use.
As shown in FIG. 6, which shows a score update chart of a particle swarm optimization process, it can be seen from FIG. 6 that the evaluation function F is evaluated when the update times gradually approach the maximum update times K 1 And taking the value, namely gradually converging the evaluation value of the historical optimal solution of the corresponding group, and completing the optimization. In an office scene, the obtained group history optimal solution corresponds to the modulation values (luminous flux) of the illumination control parameters of the lamps a to h as follows: 1021 367, 1042, 1128, 1252, 159, 429, 217 (lm).
As shown in fig. 1 to 7, the optimized lighting control parameters of each luminaire, such as the light flux modulation value, are stored in the scene lighting control parameter set 142 of the storage module 140. The image acquisition module employs a camera 14 located at a corner of the guest room. Preferably, the camera 14 may employ a wide angle lens. Preferably, the image acquisition module may also adopt a plurality of cameras to acquire images for different target areas respectively.
And (3) acquiring sample images of a user in various illumination scenes such as office, reading, receiving, entertainment and energy conservation while optimizing illumination control parameters in the illumination scenes, and forming a training image set by the sample images to train the scene detection unit.
Preferably, the identification unit in the scene detection unit may use a Support Vector Machine (SVM) of a linear kernel function as the classifier. For various lighting scene categories, the classifier is trained based on features of the functional areas, in particular human activity or location features. And forming the characteristic vector of the SVM by the position and the action of the personnel in each specific area in the hotel room image and the characteristic value of the related sensor data.
The training image set is derived from a segmented image segmented from a wide-angle image or the region image itself acquired for each different target region. The identification of the illumination scene is performed for each sample of the training image set, and the untrained recognition portion is trained based on the training image set. The structural parameters of the scene detection unit after training, such as the parameters of each SVM in the recognition section, may be stored in the scene detection unit structural parameter set.
Specific lighting scene identification is defined according to the usage characteristics of the scene and the human activity characteristics. For example, for an office scene, the scene features of which are the processing work of the user on the desk, for a scene detection recognition section of the office scene, image features of a person beside the desk may be mainly recognized, and a human body sensor such as an infrared sensor may be provided near the desk, the image features and the values of the infrared sensor together constituting a feature vector of the present scene.
In another example, for a guest meeting scene, a feature of entering a guest room by multiple persons can be obtained in a person counter at a guest room gate, and a scene image feature of gathering multiple persons in a sofa or other area can be obtained.
Preferably, the feature of the feature vector may be a static image feature at a time point or a human movement feature detected in a plurality of continuous images. For example, the user walks from the doorway to the vicinity of the desk, passes through multiple areas, and eventually stops near the desk.
Preferably, a part of the sample image of the training image set may be further left as a verification set to verify the trained scene detection unit.
Preferably, the scene detection unit includes a plurality of recognition parts, one of the plurality of recognition parts corresponding to human body detection of one local area, and each recognition part corresponding to one classifier.
After the recognition part is trained by the training image set, acquiring an image of a guest room of a business hotel and sensing data corresponding to the image, extracting image features and feature values of the sensing data, and determining the lighting scene recognition of the image according to the image features and the feature values of the sensing data.
Preferably, one or more illumination scene recognition of the image is determined according to the image characteristics of each region and the characteristic values of the corresponding sensing data; when the determined illumination scene of the image is identified as a plurality, the illumination scene of the image is determined to be identified as a composite illumination scene.
Preferably, for the comprehensive lighting scene recognition, the key area is marked for each lighting scene corresponding to each lighting scene recognition of the multiple lighting scene recognition included in the comprehensive lighting scene recognition, and the lighting fixtures in the key area of the lighting scene are regulated and controlled according to the fixture regulation parameters of the lighting scene. If the key area of the office scene is a desk area, the lighting lamp in the desk area, for example, right above the desk area, regulates and controls lamp regulation parameters according to the office scene.
Preferably, when the multiple lighting scene identifications included in the integrated lighting scene identification are intersected with the key areas of the corresponding multiple lighting scenes, the lighting fixtures at the intersected part are regulated according to the weighted average value of the fixture regulation parameters of the lighting scenes corresponding to all the lighting scene identifications with the intersected areas marked as the key areas.
Preferably, the weighted average may be an arithmetic average.
Referring to fig. 1, 5 and 7, after all the five groups of lighting scenes are optimized, the scene detection unit 133 identifies the scenes and transmits the driving current value to the driver 200 of the LED lamp 300 through the communication interface unit 152 in the output module, and the LED lamp groups formed by the LED lamps 300 respectively adjust the driving current of each LED lamp according to the lamp regulation parameters corresponding to the lighting scenes, so as to turn on the corresponding lighting scenes.
Preferably, on the basis of automatically identifying a scene, the lighting control device of the invention can also be provided with a control panel 112 in the input module 110, and a user can manually adjust the operation on the control panel 112, and after the operation is processed by the processing module, the light emitting of the LED lamp group is manually adjusted through the output module. And the adjusted results may be maintained in the centralized control host 10 for use in the next scenario. At the same time, the control panel 112 may also be used to input parameters during the optimization process.
Preferably, the control panel 112 comprises two identical touch control panels, which are respectively arranged on the hotel room hall 8 and the right side wall of the bed surface 1. The control panel 112 adopts a touch screen, an automatic scene control key is arranged on the interface of the control panel, and after clicking the key, the output module transmits a corresponding driving current value to the driver of the LED lamp according to the identified lighting scene.
The lamp regulation parameter may be a lamp luminous flux or a driving current corresponding to the lamp luminous flux, and the driving current may be an absolute value or a percentage value relative to a rated driving current value. Assuming that the luminous flux of the LED lamp is proportional to the driving current, the lamp regulation parameters obtained by optimizing different scenes are mapped to the duty ratio of the PWM wave of the driving current and stored in the scene lighting control parameter set 142. According to the identified lighting scene, the event processing unit 132 of the processing module 130 obtains the driving current PWM wave duty ratio corresponding to the scene from the scene lighting control parameter set 142, and sends an instruction to the driver 200 through the communication interface unit 152, so as to change the driving current, respectively change the power supply current of the dimmable LED lamp, and realize the switching of the preset scene.
It will be appreciated that each luminaire may have a plurality of modulation values of the illumination control parameters, e.g. brightness, color, during particle swarm optimizationTemperature, colour, etc., i.e. the particle population may be provided with a plurality of elements x ij () To represent the i groups of particle illumination control parameter modulation values of a certain lamp in the guest room.
The above-described embodiments do not limit the scope of the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the above embodiments should be included in the scope of the present invention.

Claims (10)

1. A business hotel guest room illumination control device based on scene automatic identification comprises an input module, an image acquisition module, a human body sensor, a processing module, an output module and a storage module,
the processing module comprises an optimizing processing unit, an event processing unit and a scene detection unit, wherein the storage module stores an illumination space model of a business hotel guest room, particle swarm optimizing processing parameters and scene illumination control parameters optimized by the processing module, and the optimizing processing unit is configured to:
responding to an optimized processing instruction of an event processing unit, scoring the standard reaching degree of illumination and illumination uniformity of each area of a business hotel guest room by taking the illumination standard or general requirement of the business hotel industry as a reference based on an illumination space model which is formed by describing the geometric parameters of each area in the business hotel guest room and the light distribution parameters of each lamp distributed in the guest room, establishing an evaluation function F of an illumination effect, supplementing the evaluation function F according to the scene use requirement of the business hotel illumination, and establishing a total evaluation function F by weighting and summing the F and the supplemented items; optimizing the illumination control parameters of each lamp in the guest room by adopting particle swarm optimization based on the overall evaluation function F, and mapping the optimization result into scene illumination control parameters;
The scene detection unit is configured to:
training based on a training image set, extracting image characteristics and characteristic values of sensing data based on images acquired by an image acquisition module and human body sensor sensing data corresponding to the images, determining illumination scene identification of the images according to the image characteristics and the characteristic values of the sensing data, and sending identification results to an event processing unit;
the event processing unit is configured to:
responding to the signal input of the input module, displaying the processed signal to a user through the output module, responding to the identification result signal of the scene detection unit, outputting the scene illumination control parameter corresponding to the identification result obtained by the optimization processing unit, and converting the illumination control parameter corresponding to the scene into a driving current value of the lamp;
the particle swarm optimization processing flow is as follows:
s1, randomly generating an initial set X (1) of particle groups N*n Its element x ij (1) I=1, 2,..n, j=1, 2,..n; n is the number of lamps to be optimized in the guest room, namely the dimension of the particle swarm, and N is the size of the particle swarm; each element x ij (1) The method comprises the steps that i groups of particles in a guest room are represented as initial illumination control parameter modulation values, illumination distribution of each region of the guest room in a hotel is calculated for all n lamps corresponding to the i groups of particles according to a light distribution model when each lamp illuminates according to the current illumination control parameter modulation values, and an overall evaluation function F is adopted to evaluate illumination effects corresponding to the i groups of particles;
Respectively obtaining initial grading values of the lighting control parameter modulation value vectors of each particle group, taking the initial grading values as initial grading values of self-history optimal solutions, and recording the lighting control parameter modulation value vectors;
meanwhile, the maximum value in all N groups of particle initial scoring values is recorded as the initial scoring value of the group history optimal solution, and the illumination control parameter modulation value vector is recorded, so that the updating frequency k=1;
s2, in the particle swarm optimization process, each element x ij () All corresponding to a change value v ij () Change value v ij () Modulation value x of illumination control parameter of corresponding lamp j ij () V is also needed to be carried out on the basis of (1) ij () To ensure its effectiveness, v ij () The value interval should not be greater than x ij () 10% of the maximum value that can be taken; if x is changed ij () Greater than x ij () Maximum value possible to be taken, x ij () Still taking its maximum value;
randomly generated set V (k) N*n The elements thereof respectively represent a collection X (k) N*n The values of the elements are changed according to the following formula N*n X (k) N*n Performing multiple updates:
v ij (k+1)＝wv ij (k)+c 1 r 1 (P ij (k)-x ij (k))+c 2 r 2 (G j (k)-x ij (k))
x ij (k+1)＝x ij (k)+v ij (k+1)
wherein k is the current update times; p (P) ij (k) The specific illumination control parameter modulation values corresponding to the self-history optimal solution of the ith group of particle data in the kth updating are represented, the specific illumination control parameter modulation values are obtained by comparing the score values of the updated illumination control parameter modulation values and the self-history optimal solution, and if the updated score values are larger than the score values of the self-history optimal solution, the updated illumination control parameter modulation values are recorded as new self-history optimal solutions;
G j (k) The particle group history optimal solution representing the jth specific illumination control parameter modulation value in the kth updating is obtained by comparing the updated illumination control parameter modulation values of each group with the scoring value of the group history optimal solution;
c 1 ，c 2 is constant, r 1 ，r 2 Is a random number uniformly distributed between the intervals (0, 1); w is a weight coefficient; w (w) max ，w min Respectively a maximum value and a minimum value of weight coefficients, X (k+1) N*n V (k+1) N*n Representation pair X (k) N*n V (k) N*n Is the kth optimization update of (c);
k represents the maximum number of updates, and when K equals K, the update is ended and G is output j (k) The most recorded group historyAnd the optimal solution of the illumination control parameter modulation value is used as a final output result.
2. The automated scene recognition-based lighting control device of business hotel guest room of claim 1, wherein the scene detection unit is further configured to:
determining one or more lighting scene identifications of the image according to the image features and feature values of the sensing data; when the determined illumination scene of the image is identified as a plurality, the illumination scene of the image is determined to be identified as a composite illumination scene.
3. The lighting control device for business hotel guest room based on scene automatic identification according to claim 1, wherein the optimizing processing unit further converts and stores the optimizing result as lighting scene lamp regulation parameters;
The lamp regulation and control parameter is the lamp luminous flux or the driving current corresponding to the lamp color temperature;
the magnitude of the drive current is an absolute value or a percentage value relative to a rated drive current value.
4. A business hotel guest room lighting control device based on scene automatic identification as defined in claim 3, wherein the event processing unit is further configured to:
when the lighting scene is identified as the comprehensive lighting scene identification, respectively carrying out adjustment control on the driving current of the LED lamp group according to the weighted sum of the lamp regulation parameters corresponding to the plurality of lighting scenes;
or when the lighting scene is identified as the comprehensive lighting scene identification, marking a key area on the lighting scene corresponding to each lighting scene identification of the plurality of lighting scene identifications, wherein the lighting lamps in the key area of the lighting scene are regulated and controlled according to the lamp regulation and control parameters of the lighting scene; when the key areas are intersected, the lighting fixtures at the intersected parts are regulated and controlled according to the weighted average value of the lamp regulation parameters of the lighting scenes corresponding to all the lighting scene identification of the key areas marked by the intersected areas.
5. The automated scene recognition-based lighting control device of business hotel guest room of claim 1, wherein the optimization processing unit is further configured to:
dividing the business hotel guest room into three areas, namely a working area, a nearby area and a background area, according to the use frequency of the business hotel guest room, wherein the areas with more uses, such as a bed surface, a writing desk and a sofa, are defined as working areas; other users may use the area which is not remained for a long time and is listed as a nearby area; the area between the bed surface and the wall of the toilet and the position of the windowsill are listed in the background area;
taking the illuminance and the uniformity of the illuminance of each area as main indexes, and establishing an evaluation function f in an optimization processing unit:
f＝w 1 ×u(E 1 )+w 2 ×u(E 2 )+w 3 ×u(E 3 )+w 4 ×u(U 1 )+w 5 ×u(U 2 )
wherein E is 1 For the horizontal illuminance of the working area E 2 For illuminance of adjacent area E 3 For background area illumination, U 1 For uniformity of illumination of working area, U 2 The illuminance uniformity of the adjacent area; w (w) 1 、w 2 、w 3 、w 4 、w 5 Respectively assigning weights of the indexes by using an analytic hierarchy process and inputting the weights through an input module; u () represents the proximity between the calculated value obtained from the luminaire light distribution model and the reference value for each index, for E 1 、E 2 、E 3 In other words u () =1 when the calculated value is equal to the reference value;
the analytical hierarchy process is as follows: establishing a judgment matrix A according to the number h of the weight coefficients h*h And uses interval [1,9 ]]The comparison matrix of two integers in the matrix is filled, and the numerical value of each element isRepresenting the importance degree of the index corresponding to the corresponding row weight item compared with the index corresponding to the corresponding column weight item, wherein the numerical value is a subjective evaluation result or a statistical result obtained through sample investigation; then, matrix A h*h Substituting into a geometric mean equation, determining each weight coefficient value, wherein the geometric mean equation is as follows:
wherein a is ij The importance ratio of the ith weight item to the jth weight item.
6. The automated scene recognition-based lighting control device of business hotel guest room of claim 5, wherein the optimization processing unit is further configured to: taking the lighting energy conservation of the lamp as an optimization target, supplementing the evaluation function F, wherein the energy consumption of the lamp is reduced to the minimum while the illumination and illumination uniformity indexes are ensured in each working area, and if n lamps are shared in a guest room, the new evaluation function F is provided 1 Expressed as:
where P (i) is the power consumption of the ith lamp.
7. The automated scene recognition-based lighting control device of business hotel guest room of claim 6, wherein the optimization processing unit is further configured to:
Aiming at office scenes, the illumination of a writing desk and illumination uniformity are required to meet reference conditions as far as possible, and a new evaluation function F is obtained 2 Expressed as:
wherein E is 11 Indicating the illumination of the writing desk, U 11 Indicating illumination uniformity of writing desk, q 1 、q 2 、k 1 、k 2 、k 3 The weight coefficient of the corresponding item is obtained by a hierarchical analysis method.
8. The automated scene recognition-based lighting control device of business hotel guest room of claim 6, wherein the optimization processing unit is further configured to: aiming at a guest receiving scene, the overall illumination and the overall illumination uniformity in each area are required to be more similar, the illumination reference value in each area is uniformly set to be a fixed value, the illumination uniformity reference value is set to be a fixed value, and a new evaluation function F is required 3 Expressed as:
wherein k is 1 、k 2 The weight coefficient of the corresponding item is obtained by a hierarchical analysis method.
9. The automated scene recognition-based lighting control device of business hotel guest room of claim 6, wherein the optimization processing unit is further configured to: for entertainment scenes, a lower illumination intensity of the background and the adjacent areas is required, and a new evaluation function F is obtained 4 Expressed as:
Wherein E is 12 Represents the horizontal illumination of the writing desk outside the working area, q 1 、q 2 、q 3 、k 1 、k 2 、k 3 The weight coefficient of the corresponding item is obtained by a hierarchical analysis method.
10. Root of Chinese characterThe automated scene recognition-based lighting control device of business hotel guest room of claim 6, wherein the optimization processing unit is further configured to: for leisure reading scenes, the brightness of m lamps near the head of the bed is required to be higher, and a new evaluation function F is obtained 5 Expressed as:
wherein I is t Respectively represents the brightness, q of the t-th lamp near the head of the bed t Representing their respective weight coefficients; k (k) 1 、k 2 、k 3 The values of the weight coefficients are obtained by a hierarchical analysis method for the weight coefficients of the corresponding items.
CN202110233109.XA 2018-11-02 2018-11-02 Business hotel guest room illumination control device based on scene automatic identification Active CN112969254B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110233109.XA CN112969254B (en) 2018-11-02 2018-11-02 Business hotel guest room illumination control device based on scene automatic identification

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811343930.1A CN109874198B (en) 2018-11-02 2018-11-02 Commercial hotel guest room illumination control device based on scene automatic identification
CN202110233109.XA CN112969254B (en) 2018-11-02 2018-11-02 Business hotel guest room illumination control device based on scene automatic identification

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
CN201811343930.1A Division CN109874198B (en) 2018-11-02 2018-11-02 Commercial hotel guest room illumination control device based on scene automatic identification

Publications (2)

Publication Number Publication Date
CN112969254A CN112969254A (en) 2021-06-15
CN112969254B true CN112969254B (en) 2024-02-09

Family

ID=66916958

Family Applications (2)

Application Number Title Priority Date Filing Date
CN201811343930.1A Active CN109874198B (en) 2018-11-02 2018-11-02 Commercial hotel guest room illumination control device based on scene automatic identification
CN202110233109.XA Active CN112969254B (en) 2018-11-02 2018-11-02 Business hotel guest room illumination control device based on scene automatic identification

Family Applications Before (1)

Application Number Title Priority Date Filing Date
CN201811343930.1A Active CN109874198B (en) 2018-11-02 2018-11-02 Commercial hotel guest room illumination control device based on scene automatic identification

Country Status (1)

Country Link
CN (2) CN109874198B (en)

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109874209A (en) * 2018-11-06 2019-06-11 中国计量大学 Commercial hotel guest room scene lighting system based on scene automatic identification
CN110337156A (en) * 2019-06-25 2019-10-15 广东以诺通讯有限公司 A kind of intelligent lamp
CN110381634A (en) * 2019-07-08 2019-10-25 惠州市海尼克电子科技有限公司 Illumination control method, storage medium and lighting device
CN110809353A (en) * 2019-10-28 2020-02-18 星络智能科技有限公司 Lamplight matching method, server and storage medium
CN112353549A (en) * 2020-10-14 2021-02-12 深圳中科大唐科技有限公司 Device and method for improving sleep breathing
CN112365690A (en) * 2020-10-30 2021-02-12 四川大学华西医院 Intelligent wearable device based on scene recognition and early warning method thereof
CN113063504A (en) * 2020-12-23 2021-07-02 泰州可以信息科技有限公司 Studio lighting scene recognition
CN113313061A (en) * 2021-06-18 2021-08-27 张学勇 Illumination scene recognition method for ear-nose-throat department
CN114258176A (en) * 2021-12-31 2022-03-29 欧普照明股份有限公司 Lamp and lamp control method
GB202411362D0 (en) * 2022-12-09 2024-09-18 Wgwg Pty Ltd Video display ambient lighting synchronisation

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2011222429A (en) * 2010-04-13 2011-11-04 Hakutsu Technology Corp Light control system of illumination
CN106304494A (en) * 2016-08-31 2017-01-04 华南师范大学 The LED lamp intelligent control system of a kind of scene Recognition and method
CN106961772A (en) * 2015-12-31 2017-07-18 群光电能科技股份有限公司 Instant Lighting Control Assembly and instant illumination control method

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101707052B (en) * 2009-12-09 2012-03-28 中航华东光电有限公司 RGB-based method for adjusting white balance of LED backlight
CN104811693B (en) * 2015-04-14 2016-06-29 宁波大学 A kind of stereo image vision comfort level method for objectively evaluating
CN105426669B (en) * 2015-11-10 2018-01-12 东南大学 A kind of general office energy efficiency of illumination methods of marking
CN106815470B (en) * 2016-12-27 2019-04-23 浙江工业大学 A kind of appraisal procedure that office scene lighting environment influences Vision Health

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2011222429A (en) * 2010-04-13 2011-11-04 Hakutsu Technology Corp Light control system of illumination
CN106961772A (en) * 2015-12-31 2017-07-18 群光电能科技股份有限公司 Instant Lighting Control Assembly and instant illumination control method
CN106304494A (en) * 2016-08-31 2017-01-04 华南师范大学 The LED lamp intelligent control system of a kind of scene Recognition and method

Also Published As

Publication number Publication date
CN112969254A (en) 2021-06-15
CN109874198A (en) 2019-06-11
CN109874198B (en) 2021-03-12

Similar Documents

Publication Publication Date Title
CN112969254B (en) Business hotel guest room illumination control device based on scene automatic identification
CN112584572B (en) Open Office Lighting System
CN112533319B (en) Scene type classroom intelligent lighting control device
CN110035578B (en) Open type office lighting system based on mixed lighting
CN109874209A (en) Commercial hotel guest room scene lighting system based on scene automatic identification
CN109874197B (en) Commercial hotel guest room illumination control method based on scene automatic identification
CN109874202A (en) The adaptive lighting system of kindergarten one classroom scene type, control device and control method
CN108738215A (en) A kind of automatic adjustment desk lamp technical parameter method, apparatus and desk lamp
CN106340071A (en) Intelligent classroom system based on cloud computing
CN109872020B (en) Lighting control device for business hotel guest room
KR100946202B1 (en) Sensitive illumination system
CN109874217B (en) Public dormitory mixed lighting method based on anisotropic symmetrical ceiling lamp
CN110139449A (en) A kind of full room lighting system of intelligence based on human body attitude identification
CN110062498A (en) Public Quarters blending illumination system, method and optimization method based on the controllable ceiling lamp of subregion
CN118042684A (en) Light control method and system based on artificial intelligence
CN112689353B (en) Lighting control method for guest room of business hotel
CN112218414A (en) Method and system for adjusting brightness of self-adaptive equipment
CN116661329B (en) Cooperative regulation and control method for intelligent classroom acousto-optic system
CN116939921B (en) Intelligent household lighting control system and method based on Internet of things
CN109874199A (en) A kind of commercial hotel guest room scene lighting system
CN116209112A (en) LED ceiling lamp based on gesture estimation and automatic scene recognition
Baviskar Artificial Intelligence in Lighting for Educational Institutions
CN118591064A (en) Intelligent illumination adjusting system and method based on big data
CN115866855A (en) Intelligent dimming down lamp control system

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
GR01 Patent grant
TR01 Transfer of patent right

Effective date of registration: 20240703

Address after: 230000 B-1015, wo Yuan Garden, 81 Ganquan Road, Shushan District, Hefei, Anhui.

Patentee after: HEFEI MINGLONG ELECTRONIC TECHNOLOGY Co.,Ltd.

Country or region after: China

Address before: 310018, No. 258, source street, Xiasha Higher Education Park, Hangzhou, Zhejiang

Patentee before: China Jiliang University

Country or region before: China

TR01 Transfer of patent right