CN116528438A

CN116528438A - Intelligent dimming method and device for lamp

Info

Publication number: CN116528438A
Application number: CN202310481444.0A
Authority: CN
Inventors: 朱秉锋; 陈谦
Original assignee: Guangzhou Leemc Lighting Technology Co ltd
Current assignee: Guangzhou Leemc Lighting Technology Co ltd
Priority date: 2023-04-28
Filing date: 2023-04-28
Publication date: 2023-08-01
Anticipated expiration: 2043-04-28
Also published as: CN116528438B

Abstract

The invention provides an intelligent dimming method and device of a lamp, wherein the method comprises the following steps: the method comprises the steps of obtaining sounds of a speaker in a plurality of conference rooms in t time periods, extracting the change rule of emotion of the speaker according to time sequence, detecting emotion of the speaker, obtaining target emotion scores of the speaker, and adjusting brightness and color of the lamp according to the target emotion scores. The invention has the beneficial effects that: different colors are adjusted according to emotion of a speaker, so that participants can intuitively feel the color, and participation of the participants in a conference is improved to a certain extent.

Description

Intelligent dimming method and device for lamp

Technical Field

The invention relates to the field of artificial intelligence, in particular to an intelligent dimming method and device for a lamp.

Background

Color is used as a visual signal in early processing of the visual signal and always depends on an important variable which is considered to influence emotion, when a conference is carried out, as the color of lamplight is generally constant, the participation will of participants in the conference is not strong, and the talker who hosts the conference generally needs to show different attitudes, and when the conference time is too long, the participation of the participants is not high, so that the participation of the participants in the conference needs to be improved.

Disclosure of Invention

The invention mainly aims to provide an intelligent dimming method and device for a lamp, and aims to solve the problem that the participation degree of the existing participants in a conference is not high.

The invention provides an intelligent dimming method of a lamp, which comprises the following steps:

acquiring the sound of a speaker in a plurality of conference rooms in t time periods and emotion scores in the t+1th time period;

carrying out emotion analysis on the sound in each time period to obtain emotion vectors of the speaker corresponding to each time period;

calculating the feature vector X of each time period according to the emotion vector _i ＝

{(Q _t，t-1 )，(Q _t，t-2 )，...，(Q _t，z )...，(Q _t，2 )，(Q _t，1 ) T represents a t-th period, X _i Feature vector representing the i-th conference room, (Q) _t，z ) The change rate of the emotion vector between the t time period and the z time period is represented, t, z and i belong to positive integers, and t is more than z, i is less than or equal to n;

carrying out standardization processing on each feature vector according to a preset method to obtain a standard data set;

dividing the standard data set into a training data set and a test data set according to a preset proportion;

inputting the training data set and each emotion score of the conference room corresponding to the training data set into a preset model, and training the preset model according to an optimal super-parameter method;

Detecting the trained model through the test data set and the corresponding emotion scores, and obtaining a target model when the detection result meets the training requirement of the model;

collecting the sound of a target speaker in a conference room in the current time period, and converting the sound into a corresponding target emotion vector;

calculating a target feature vector of a target speaker according to the target emotion vector, and inputting the target emotion vector into the target model to obtain a target emotion score of the target speaker;

and adjusting the brightness and the color of the lamp based on the target emotion score.

Further, before the step of inputting the training data set and the emotion scores of the meeting rooms corresponding to the training data set into a preset model and training the preset model according to the optimal super-parameter method, the method further comprises:

acquiring an initial model of a plurality of different super-parameter combinations;

dividing the training data set into a training set and a verification set according to a preset proportion;

inputting the training set and each emotion score of the corresponding conference room into each initial model for training to obtain a plurality of corresponding temporary models;

verifying each temporary model through the verification set to obtain a verification result of each temporary model;

And selecting the temporary model with the optimal verification result as the preset model based on the verification result.

Further, the light emitting element of the lamp is an LED lamp, and the step of adjusting the brightness and color of the lamp based on the target emotion score includes:

acquiring required light brightness and color based on the target emotion score;

and setting parameters of the LED lamp according to the required brightness and color of the lamp, thereby realizing the brightness and color of the lamp.

Further, the step of performing normalization processing on each feature vector according to a preset method to obtain a standard data set includes:

extracting maximum value points and minimum value points in each characteristic vector, and arranging according to a time sequence to obtain an extremum sequence;

fitting the extremum sequence by using a cubic spline interpolation function to obtain an upper envelope X and a lower envelope X _max(t) and X_min (t)；

Taking the average value of the upper envelope curve and the lower envelope curve as an envelope curve average value m (t); wherein,

subtracting the envelope mean value from the feature vector to obtain a target sequence;

judging whether the target sequence passes through an intrinsic mode function test;

if the detection is passed, the target sequence is marked as a random oscillation function, otherwise, the target sequence is marked as a first eigenvector and the target sequence is recalculated until the random oscillation function is obtained;

Subtracting the first characteristic vector from the characteristic vector to obtain a second characteristic vector, and repeating to obtain a plurality of random oscillation functions until the calculated envelope is symmetrical and the average value of the envelope is 0, so as to obtain a plurality of random oscillation functions;

and collecting random oscillation functions of each feature vector, so as to obtain standard data corresponding to each feature vector, and further obtain a standard data set consisting of all the standard data.

Further, the step of performing emotion analysis on the sound in each time period to obtain emotion vectors of the speaker corresponding to each time period includes:

extracting text information and voice parameter information corresponding to the voice in each time period;

performing text emotion analysis on the text information to obtain first emotion information expressed by the text information, and performing voice emotion analysis on the voice parameter information to obtain second emotion information expressed by the voice parameter information;

and obtaining emotion vectors corresponding to each time period based on the first emotion information and the second emotion information.

The invention also provides an intelligent dimming device of the lamp, which comprises:

The acquisition module is used for acquiring the sound of the speaker in the plurality of conference rooms in t time periods and the emotion scores in the t+1th time period;

the analysis module is used for carrying out emotion analysis on the sound in each time period to obtain emotion vectors of the speaker corresponding to each time period;

a first calculation module for calculating the feature vector of each time period according to the emotion vector

X _i ＝

the processing module is used for carrying out standardized processing on each characteristic vector according to a preset method to obtain a standard data set;

the distribution module is used for dividing the standard data set into a training data set and a test data set according to a preset proportion;

the input module is used for inputting the training data set and each emotion score of the meeting room corresponding to the training data set into a preset model, and training the preset model according to an optimal super-parameter method;

the detection module is used for detecting the trained model through the test data set and the corresponding emotion scores, and when the detection result meets the training requirement of the model, a target model is obtained;

The collecting module is used for collecting the sound of a target speaker in the conference room in the current time period and converting the sound into a corresponding target emotion vector;

the second calculation module is used for calculating a target feature vector of a target speaker according to the target emotion vector, and inputting the target emotion vector into the target model to obtain a target emotion score of the target speaker;

and the adjusting module is used for adjusting the brightness and the color of the lamp based on the target emotion score.

Further, the intelligent dimming device of the lamp further comprises:

the initial model acquisition module is used for acquiring a plurality of initial models of different super parameter combinations;

the data set distribution module is used for dividing the training data set into a training set and a verification set according to a preset proportion;

the emotion score input module is used for inputting the training set and each emotion score of the corresponding conference room into each initial model for training to obtain a plurality of corresponding temporary models;

the verification module is used for verifying each temporary model through the verification set to obtain a verification result of each temporary model;

and the module is used for selecting the temporary model with the optimal verification result as the preset model based on the verification result.

Further, the light emitting element of the lamp is an LED lamp, and the adjusting module includes:

the acquisition sub-module is used for acquiring the required light brightness and color based on the target emotion score;

and the setting submodule is used for setting parameters of the LED lamp according to the required brightness and color of the lamp, so that the brightness and color of the lamp are realized.

Further, the processing module includes:

the extraction submodule is used for extracting maximum value points and minimum value points in each characteristic vector and obtaining an extremum sequence according to time sequence arrangement;

a fitting sub-module for fitting the extremum sequence by using a cubic spline interpolation function to obtain upper and lower envelope X _max(t) and X_min (t)；

The value taking sub-module is used for taking the average value of the upper envelope curve and the lower envelope curve as the average value m (t) of the envelope curve; wherein,

the first calculation sub-module is used for subtracting the envelope mean value from the feature vector to obtain a target sequence;

the judging submodule is used for judging whether the target sequence passes through the natural mode function test;

the marking sub-module is used for marking the target sequence as a random oscillation function if the detection is passed, otherwise marking the target sequence as a first characteristic vector and recalculating the target sequence until the random oscillation function is obtained;

The second calculation sub-module is used for subtracting the first characteristic vector from the characteristic vector to obtain a second characteristic vector, and repeatedly obtaining a plurality of random oscillation functions until the calculated envelope is symmetrical and the average value of the envelope is 0, so that a plurality of random oscillation functions are obtained;

and the aggregation sub-module is used for aggregating the random oscillation functions of each feature vector so as to obtain standard data corresponding to each feature vector and further obtain a standard data set consisting of all the standard data.

Further, the analysis module includes:

the information extraction sub-module is used for extracting text information and voice parameter information corresponding to the voice in each time period;

the emotion analysis submodule is used for carrying out text emotion analysis on the text information to obtain first emotion information expressed by the text information, and carrying out voice emotion analysis on the voice parameter information to obtain second emotion information expressed by the voice parameter information;

and the emotion vector calculation operator module is used for obtaining emotion vectors corresponding to each time period based on the first emotion information and the second emotion information.

The invention has the beneficial effects that: the method has the advantages that the voice of the speaker in the plurality of conference rooms in t time periods is obtained, so that the change rule of the emotion of the speaker is extracted according to time sequence, the emotion of the speaker is detected, the target emotion score of the speaker is obtained, the brightness and the color of the lamp are adjusted according to the target emotion score, different colors are adjusted according to the emotion of the speaker, the participant can intuitively feel the voice, and the participation degree of the participant in the conference is improved to a certain extent.

Drawings

Fig. 1 is a flow chart of a smart dimming method of a lamp according to an embodiment of the invention;

fig. 2 is a schematic block diagram of a smart dimming device of a lamp according to an embodiment of the present invention;

fig. 3 is a block diagram schematically illustrating a structure of a computer device according to an embodiment of the present application.

The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

It should be noted that, in the embodiments of the present invention, all directional indicators (such as up, down, left, right, front, and back) are merely used to explain the relative positional relationship, movement conditions, and the like between the components in a specific posture (as shown in the drawings), if the specific posture is changed, the directional indicators correspondingly change, and the connection may be a direct connection or an indirect connection.

The term "and/or" is herein merely an association relation describing an associated object, meaning that there may be three relations, e.g., a and B, may represent: a exists alone, A and B exist together, and B exists alone.

Furthermore, descriptions such as those referred to as "first," "second," and the like, are provided for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implying an order of magnitude of the indicated technical features in the present disclosure. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In addition, the technical solutions of the embodiments may be combined with each other, but it is necessary to base that the technical solutions can be realized by those skilled in the art, and when the technical solutions are contradictory or cannot be realized, the combination of the technical solutions should be considered to be absent and not within the scope of protection claimed in the present invention.

Referring to fig. 1, the invention provides an intelligent dimming method of a lamp, which comprises the following steps:

s1: acquiring the sound of a speaker in a plurality of conference rooms in t time periods and emotion scores in the t+1th time period;

s2: carrying out emotion analysis on the sound in each time period to obtain emotion vectors of the speaker corresponding to each time period;

S3: calculating the feature vector X of each time period according to the emotion vector _i ＝{(Q _t，t-1 )，(Q _t，t-2 )，...，(Q _t，z )...，(Q _t，2 )，(Q _t，1 ) T represents a t-th period, X _i Feature vector representing the i-th conference room, (Q) _t，z ) The change rate of the emotion vector between the t time period and the z time period is represented, t, z and i belong to positive integers, and t is more than z, i is less than or equal to n;

s4: carrying out standardization processing on each feature vector according to a preset method to obtain a standard data set;

s5: dividing the standard data set into a training data set and a test data set according to a preset proportion;

s6: inputting the training data set and each emotion score of the conference room corresponding to the training data set into a preset model, and training the preset model according to an optimal super-parameter method;

s7: detecting the trained model through the test data set and the corresponding emotion scores, and obtaining a target model when the detection result meets the training requirement of the model;

s8: collecting the sound of a target speaker in a conference room in the current time period, and converting the sound into a corresponding target emotion vector;

s9: calculating a target feature vector of a target speaker according to the target emotion vector, and inputting the target emotion vector into the target model to obtain a target emotion score of the target speaker;

S10: and adjusting the brightness and the color of the lamp based on the target emotion score.

As described in step S1, the voices of the speaker in t time periods and emotion scores in the t+1th time period are obtained in the multiple conference rooms, wherein the t time periods of the speaker are continuous time periods, the t+1th time period is a continuous time period after the t time periods, that is, the t+1th time periods are all connected time periods, the continuous speaking is performed in the t+1th time period, the emotion scores in the t+1th time period can be manually set, and it is noted that the acquired voices are acquired in advance, that is, the conference voices which have been ended, the acquisition mode can be acquired by using a microphone, and the emotion scores are specifically scores manually set according to actual conditions. The emotion is divided into a plurality of basic major classes such as happiness, sadness, surprise, fear and the like based on the independent emotion, so that each emotion score corresponds to one basic major class.

As described in step S2, emotion analysis is performed on the sound in each time period to obtain emotion vectors of the speaker corresponding to each time period, specifically, the analysis may be performed by obtaining a text corresponding to the sound, and then identifying the text, or performing analysis according to the language of the sound, or a combination of the two.

As described in the above step S3, the feature vector of each time period is calculated according to the emotion vector, specifically, it may be defined that the emotion vector of each moment is related to the emotion vector in the first 1 time period, so as to obtain the feature vector. I.e. X _i ＝

as described in step S4, the feature vectors are standardized according to a preset method to obtain a standard data set, where the standardized processing may be to reduce the dimension of the data and remove unnecessary data, for example, a sliding window manner, or a manner of reducing the dimension, converting the data, etc., which will not be described in detail herein.

And (5) dividing the standard data set into a training data set and a test data set according to a preset proportion, inputting the training data set and each emotion score of the training data set corresponding to the conference room into a preset model, training the preset model according to an optimal super-parameter method, detecting the trained model through the test data set and the corresponding emotion score, and obtaining a target model when the detection result meets the training requirement of the model. The model may be an SVM model or an ANN model, where an SVM (support vector machine) is a most popular machine learning technology, is an approximation of structural risk minimization, seeks an optimal combining point between generalization performance and fitting performance of the model, and is not a traditional empirical risk minimization, and skillfully solves the dimension problem, and may process standardized data, specifically, a training method is the same as an existing SVM training method, and is not repeated herein, and the preset SVM model is a preset SVM model.

And S8-S10, collecting the sound of a target speaker in a conference room in the current time period, converting the sound into a corresponding target emotion vector, calculating a target feature vector of the target speaker according to the target emotion vector, inputting the target emotion vector into the target model to obtain a target emotion score of the target speaker, and adjusting the brightness and color of the lamp based on the target emotion score, wherein the obtained sound of the target speaker is converted into the target feature vector to serve as the input of the target model, the target emotion score is obtained, the emotion of the corresponding target speaker can be obtained according to the target emotion score, and then the brightness and color corresponding to the lamp are adjusted according to the emotion, wherein different emotions correspond to different brightness and colors. Therefore, different colors can be adjusted according to the emotion of the talker, so that the talker has visual feeling, and the participation degree of the talker to the conference is improved to a certain extent.

In one embodiment, before step S6 of inputting the training data set and the emotion scores of the meeting rooms corresponding to the training data set into a preset model and training the preset model according to the optimal hyper-parameter method, the method further includes:

S501: acquiring an initial model of a plurality of different super-parameter combinations;

s502: dividing the training data set into a training set and a verification set according to a preset proportion;

s503: inputting the training set and each emotion score of the corresponding conference room into each initial model for training to obtain a plurality of corresponding temporary models;

s504: verifying each temporary model through the verification set to obtain a verification result of each temporary model;

s505: and selecting the temporary model with the optimal verification result as the preset model based on the verification result.

As described in the above steps S501-S505, the selection of the model is achieved, and since the generalization ability of the model depends largely on the original parameters, for example, the generalization ability of the SVM model depends largely on the penalty coefficient, the insensitive loss parameter, and the kernel function width parameter. The confidence coefficient is used for controlling the exceeding sample punishment degree, the confidence range and experience risk proportion of the learning machine are adjusted, for example, the model training is difficult due to overlarge punishment coefficient, the model under fitting is easy to be caused due to overlarge punishment coefficient, the magnitude of regression error of the decision function is mainly controlled by the confidence range, the quantity of support vectors is too small due to overlarge punishment coefficient, and the prediction accuracy is reduced; the experience risk proportion influences the complexity degree of the distribution of the sample in the high-dimensional feature space, so that a plurality of initial SVM models with different hyper-parameter combinations can be used, the temporary SVM model with the optimal verification result is selected as the preset SVM model, and the prediction accuracy of the model is further improved.

In one embodiment, the light emitting element of the light fixture is an LED light, and the step S10 of adjusting the brightness and color of the light fixture based on the target emotion score includes:

s1001: acquiring required light brightness and color based on the target emotion score;

s1002: and setting parameters of the LED lamp according to the required brightness and color of the lamp, thereby realizing the brightness and color of the lamp.

As described in the above steps S1001-S1002, the required light brightness and color are obtained based on the target emotion score; the corresponding relation between the emotion score and the brightness and the color of the light can be preset, the required brightness and the color of the light can be directly obtained after the target emotion score is obtained, and then the parameters of the LED lamp are set according to the required brightness and the color of the light, so that the brightness and the color of the lamp are realized.

In one embodiment, the step S4 of performing normalization processing on each feature vector according to a preset method to obtain a standard data set includes:

s401: extracting maximum value points and minimum value points in each characteristic vector, and arranging according to a time sequence to obtain an extremum sequence;

s402: fitting the extremum sequence by using a cubic spline interpolation function to obtain an upper envelope X and a lower envelope X _max(t) and X_min (t)；

S403: taking the average value of the upper envelope curve and the lower envelope curve as an envelope curve average value m (t); wherein,

s404: subtracting the envelope mean value from the feature vector to obtain a target sequence;

s405: judging whether the target sequence passes through an intrinsic mode function test;

s406: if the detection is passed, the target sequence is marked as a random oscillation function, otherwise, the target sequence is marked as a first eigenvector and the target sequence is recalculated until the random oscillation function is obtained;

s407: subtracting the first characteristic vector from the characteristic vector to obtain a second characteristic vector, and repeating to obtain a plurality of random oscillation functions until the calculated envelope is symmetrical and the average value of the envelope is 0, so as to obtain a plurality of random oscillation functions;

s408: and collecting random oscillation functions of each feature vector, so as to obtain standard data corresponding to each feature vector, and further obtain a standard data set consisting of all the standard data.

As described in the above steps S401-S408, in order to improve the judging capability of the model, implicit features may be extracted, specifically, feature vectors may be decomposed, minimum value points and maximum value points in each of the feature vectors are extracted, the extremum sequences are obtained by arranging according to time sequence, and then the extremum sequences are fitted by using a cubic spline interpolation function to obtain upper and lower envelopes X _max(t) and X_min (t) marking the average value of the upper envelope curve and the lower envelope curve as an envelope curve average value m (t); wherein,subtracting the envelope mean value from the feature vector to obtain a target sequence, if the target sequence does not pass through the natural mode function test, marking the target sequence as a first feature vector and recalculating the target sequence until the random oscillation function is obtained, if the target sequence passes through the test, marking the target sequence as the random oscillation function, and collecting the random oscillation function of each feature vector to obtain standard data corresponding to each feature vector, thereby obtaining a standard data set consisting of all standard data. The natural mode function is a preset function, can be set manually, and can improve the prediction performance of the machine learning model through the decomposition of the feature vector.

Cubic Spline interpolation (Cubic Spline Interpolation) is simply Spline interpolation, which is a process of obtaining a curve function set by solving a three-bending moment equation set mathematically through a smooth curve of a series of shape value points.

In one embodiment, the step S2 of performing emotion analysis on the sound in each time period to obtain emotion vectors of the speaker corresponding to each time period includes:

S201: extracting text information and voice parameter information corresponding to the voice in each time period;

s202: performing text emotion analysis on the text information to obtain first emotion information expressed by the text information, and performing voice emotion analysis on the voice parameter information to obtain second emotion information expressed by the voice parameter information;

s203: and obtaining emotion vectors corresponding to each time period based on the first emotion information and the second emotion information.

As described in the above steps S201 to S203, the text information includes the content expressed by the sound signal, and the speaker speaks a sentence: the eight words "I want to speak XXX today" and "I want to speak XXX today" can be text information of the sound signal. And the voice parameter information includes a voice speed, a signal-to-noise ratio, a voice size, a pitch, an average pitch, a pitch range, a pitch variation, and the like of the voice signal. The text information can be subjected to emotion analysis by utilizing an LSTM algorithm to obtain probability values of all emotion information expressed by the text information, the probability values are used as first emotion information expressed by the text information, voice emotion analysis is performed on the voice parameters by utilizing a CNN (Convolutional Neural Network) algorithm to obtain second emotion information, and emotion vectors corresponding to all time periods are obtained based on the first emotion information and the second emotion information.

Referring to fig. 2, the present invention further provides an intelligent dimming device of a lamp, including:

an acquiring module 10, configured to acquire the voices of speakers in a plurality of conference rooms in t time periods, and emotion scores in a t+1th time period;

the analysis module 20 is configured to perform emotion analysis on the sound in each time period to obtain emotion vectors of the speaker corresponding to each time period;

a first calculation module 30 for calculating the feature vector X of each time period according to the emotion vector _i ＝

the processing module 40 is configured to perform normalization processing on each feature vector according to a preset method, so as to obtain a standard data set;

an allocation module 50, configured to divide the standard data set into a training data set and a test data set according to a preset ratio;

the input module 60 is configured to input the training data set and each emotion score of the conference room corresponding to the training data set into a preset model, and train the preset model according to an optimal super-parameter method;

The detection module 70 is configured to detect the trained model according to the test data set and the corresponding emotion score, and obtain a target model when the detection result meets the training requirement of the model;

the collection module 80 is configured to collect a sound of a target speaker in the conference room in a current time period, and convert the sound into a corresponding target emotion vector;

a second calculation module 90, configured to calculate a target feature vector of a target speaker according to the target emotion vector, and input the target emotion vector into the target model, so as to obtain a target emotion score of the target speaker;

the adjusting module 100 is configured to adjust the brightness and color of the lamp based on the target emotion score.

In one embodiment, the intelligent dimming device of the lamp further comprises:

In one embodiment, the light emitting element of the lamp is an LED lamp, and the adjusting module 100 includes:

In one embodiment, the processing module 40 includes:

In one embodiment, the analysis module 20 includes:

Referring to fig. 3, a computer device is further provided in the embodiment of the present application, where the computer device may be a server, and the internal structure of the computer device may be as shown in fig. 3. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the computer is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for storing various emotion vectors and the like. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program, when executed by the processor, can implement the intelligent dimming method of the lamp according to any of the embodiments.

Those skilled in the art will appreciate that the architecture shown in fig. 3 is merely a block diagram of a portion of the architecture in connection with the present application and is not intended to limit the computer device to which the present application is applied.

The embodiment of the application also provides a computer readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the intelligent dimming method of the lamp according to any one of the above embodiments can be implemented.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium provided herein and used in embodiments may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Syncline) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.

It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, apparatus, article or method that comprises the element.

The embodiment of the application can acquire and process the related data based on the artificial intelligence technology. Among these, artificial intelligence (Artificial Intelligence, AI) is the theory, method, technique and application system that uses a digital computer or a digital computer-controlled machine to simulate, extend and extend human intelligence, sense the environment, acquire knowledge and use knowledge to obtain optimal results.

Artificial intelligence infrastructure technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and other directions.

The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.

Claims

1. The intelligent dimming method of the lamp is characterized by comprising the following steps of:

2. The intelligent dimming method of a lamp according to claim 1, wherein the step of inputting the training data set and the emotion scores of the training data set corresponding to the conference room into a preset model and training the preset model according to an optimal super-parameter method further comprises:

3. The intelligent dimming method of a lamp according to claim 1, wherein the light emitting element of the lamp is an LED lamp, and the step of adjusting the brightness and color of the lamp based on the target emotion score comprises:

4. The intelligent dimming method of a lamp as claimed in claim 1, wherein the step of performing normalization processing on each of the feature vectors according to a preset method to obtain a standard data set comprises:

5. The intelligent dimming method of a lamp as set forth in claim 1, wherein the step of performing emotion analysis on the sound in each time period to obtain emotion vectors of the speaker corresponding to each time period comprises:

6. An intelligent dimming device of a lamp, comprising:

X _i ＝

{(Q _t，t-1 )，(Q _t，t-2 )，...，(Q _t，z )...，(Q _t，2 )，(Q _t，1 ) T represents a t-th period, X _i Feature vector representing the i-th conference room, (Q) _t，z ) Representing the change rate of emotion vector between the t time period and the z time period, wherein t, z and i are positive integerst＞z，i≤n；

7. The intelligent dimming apparatus of a light fixture as recited in claim 6, wherein the intelligent dimming apparatus of a light fixture further comprises:

8. The intelligent dimming apparatus of a lamp as set forth in claim 6, wherein the light emitting element of the lamp is an LED lamp, and the adjustment module comprises:

9. The intelligent dimming apparatus of a lamp as recited in claim 6, wherein the processing module comprises:

10. The intelligent dimming apparatus of a luminaire of claim 6, wherein the analysis module comprises: