Smart home scene generation method and device, smart home scene setting method and device, equipment and storage medium

Abstract

The embodiment of the invention relates to the field of smart home, and discloses a smart home scene generation and setting method and device, equipment and a storage medium, and the method comprises the steps: forming a comprehensive state vector used for describing the comprehensive state of each piece of equipment based on the obtained equipment state information of each piece of equipment in a home environment at a first moment; performing scene classification on the comprehensive state vector by adopting a pre-trained scene classifier to obtain a scene type to which the home environment belongs at the first moment; and carrying out associated storage on the equipment state information of each equipment in the home environment at the first moment and the scene name of the scene type to which the home environment at the first moment belongs, so as to generate a scene mode corresponding to the scene type. According to the invention, the scene mode for reproducing the home environment at the first moment can be automatically generated.

Description

Technical Field

[0001] The present invention relates to the field of smart home, and more particularly to a method, device, equipment and storage medium for generating and setting a smart home scene. Background Art

[0002] With the increasing popularity of smart home products, there are more and more smart home appliances, lights and other devices that support smart control in users' homes. The control of a single device can no longer meet the user's needs for convenient control, so scene-based control is becoming more and more important. However, before performing scene control, users need to complete complex scene settings through mini-programs and apps, or complete scene settings through complex voice interactions, which is difficult for ordinary users to operate, and there is also the problem that the set scene deviates greatly from the user's expectations. Summary of the invention

[0003] The purpose of the present invention is to at least provide a smart home scene generation method, device, equipment and storage medium, which can at least solve the problem that users need to complete complex scene settings before scene control, which makes it difficult for many ordinary people to operate. The present application can automatically generate a scene mode corresponding to the home environment at the current moment when the user wants to generate the home environment at the current moment.

[0004] To solve the above technical problems, at least one embodiment of the present application provides a smart home scene generation method, including:

[0005] Based on the acquired device state information of each device in the home environment at the first moment, a comprehensive state vector is formed for describing the comprehensive state of each device;

[0006] Using a pre-trained scene classifier to perform scene classification on the comprehensive state vector to obtain the scene type to which the home environment at the first moment belongs;

[0007] The device status information of each device in the home environment at the first moment is associated with the scene name of the scene type to which the home environment at the first moment belongs and stored to generate a scene mode corresponding to the scene type.

[0008] In some optional embodiments, the scene classifier includes a plurality of preset output scene types;

[0009] The step of performing scene classification on the comprehensive state vector using a pre-trained scene classifier to obtain the scene type to which the home environment at the first moment belongs includes:

[0010] Classifying the comprehensive state vector using a pre-trained scene classifier to obtain a classification probability value of each preset scene type to which the home environment belongs at the first moment;

[0011] The scene type with the largest probability value is selected from each of the classification probability values ​​as the scene type to which the home environment at the first moment belongs.

[0012] In some optional embodiments, the forming of a comprehensive state vector for describing the comprehensive state of each device based on the acquired device state information of each device in the home environment at the first moment includes:

[0013] Encode the device status information of each device in the home environment at the first moment to obtain a status value of each device in multiple states;

[0014] Writing the state values ​​of each device under the multiple states into a preset data model to generate a device state matrix, wherein each row or column in the device state matrix represents a device state vector of a device;

[0015] The device state vectors in the device state matrix are sequentially concatenated head to tail to obtain the comprehensive state vector.

[0016] In some optional embodiments, the performing scene classification on the comprehensive state vector using a pre-trained scene classifier to obtain the scene type to which the home environment at the first moment belongs includes:

[0017] Performing feature extraction on the comprehensive state vector to obtain a high-dimensional feature vector of the comprehensive device state vector;

[0018] The high-dimensional feature vector is input into a scene classifier for scene classification to obtain the scene type to which the home environment belongs at the first moment.

[0019] In some optional embodiments, extracting features from the integrated state vector to obtain a high-dimensional feature vector of the integrated device state vector includes:

[0020] The comprehensive device state vector is subjected to feature extraction through a deep learning network to obtain the high-dimensional feature vector.

[0021] At least one embodiment of the present application further provides a smart home scene setting method, including:

[0022] Based on a given scene name, a target scene mode corresponding to the scene name is determined from a plurality of preset scene modes, and target device state information of each device included in the target scene mode is extracted; the plurality of preset scene modes are generated by the above-mentioned smart home scene generation method;

[0023] Control the device status of each device in the current home environment to be consistent with the target device status information.

[0024] At least one embodiment of the present application further provides a smart home scene generation device, including:

[0025] A processing module, configured to form a comprehensive state vector for describing a comprehensive state of each device based on the acquired device state information of each device in the home environment at the first moment;

[0026] A scene classification module, used to classify the comprehensive state vector using a pre-trained scene classifier to obtain the scene type to which the home environment belongs at the first moment;

[0027] The scene generation module is used to associate and store the device status information of each device in the home environment at the first moment with the scene name of the scene type to which the home environment belongs at the first moment, so as to generate a scene mode corresponding to the scene type.

[0028] At least one embodiment of the present application further provides a smart home scene setting device, including:

[0029] An extraction module, configured to determine, based on a given scene name, a target scene mode corresponding to the scene name from a plurality of preset scene modes, and extract target device state information of each device included in the target scene mode; the plurality of preset scene modes are generated by the above-mentioned smart home scene generation method;

[0030] The control module is used to control the device status of each device in the current home environment to be consistent with the target device status information.

[0031] At least one embodiment of the present application also provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned smart home scene generation method, or implement the above-mentioned smart home scene setting method.

[0032] At least one embodiment of the present application further provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned smart home scene generation method, or implements the above-mentioned smart home scene setting method.

[0033] The embodiments of the present application provide a method, apparatus, device and storage medium for generating and setting a smart home scene. Based on the device status information of each device in the home environment at the first moment, a comprehensive state vector for describing the comprehensive state of each device is formed; then the comprehensive state vector is subjected to scene classification using a pre-trained scene classifier to obtain the scene type to which the home environment belongs at the first moment; finally, the device status information of each device in the home environment at the first moment is associated and stored with the scene name of the scene type to which the home environment belongs at the first moment, so as to generate a scene mode corresponding to the scene type. Thus, the scene mode for reproducing the home environment at the first moment is automatically generated. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] One or more embodiments are exemplarily described by the pictures in the corresponding drawings, and these exemplary descriptions do not constitute limitations on the embodiments.

[0035] Figure 1 is a flow chart of a smart home scene generation method provided by an embodiment of the present application;

[0036] Figure 2 is a schematic diagram of a data model provided by an embodiment of the present application;

[0037] Figure 3 is a schematic diagram of a multi-layer neural network structure used by a scene classifier provided in one embodiment of the present application;

[0038] Figure 4 is a flowchart of a smart home scene setting method provided by another embodiment of the present application;

[0039] Figure 5 is a structural diagram of a smart home scene generation device provided by another embodiment of the present application;

[0040] Figure 6 It is a structural diagram of a smart home scene setting device provided by another embodiment of the present application. DETAILED DESCRIPTION

[0041] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the embodiments of the present application will be described in detail below in conjunction with the accompanying drawings. However, it will be appreciated by those skilled in the art that in the present application, many technical details are proposed in order to enable the reader to better understand the present application. However, even without these technical details and various changes and modifications based on the following embodiments, the technical scheme claimed in the present application can also be implemented. The division of the following embodiments is for the convenience of description, and the specific implementation of the present application should not be construed as any limitation, and the various embodiments can be combined and referenced with each other under the premise of no contradiction.

[0042] In order to facilitate understanding of the embodiments of the present application, relevant content about smart home scene setting is first introduced here.

[0043] With the increasing popularity of smart home products, there are more and more smart home appliances, lights and other devices that support smart control in users' homes. The control of a single device can no longer meet the user's needs for convenient control, so scene-based control is becoming more and more important. However, before performing scene control, users need to complete complex scene settings through mini-programs and apps, or complete scene settings through complex voice interactions, which is difficult for ordinary users to operate, and there is also the problem that the set scene deviates greatly from the user's expectations.

[0044] In order to solve the above-mentioned technical problem that users need to complete complex scene settings before scene control, which makes it difficult for many ordinary people to operate, the present invention proposes a smart home scene generation method. The implementation details of the smart home scene generation method of this embodiment are specifically described below. The following content is only the implementation details provided for the convenience of understanding and is not necessary for the implementation of this solution.

[0045] Embodiment 1:

[0046] The smart home scene generation method of this embodiment can be applied to electronic devices with communication, computing and data storage capabilities. The specific process can be as follows: Figure 1 As shown, including:

[0047] Step S11, based on the acquired device status information of each device in the home environment at the first moment, a comprehensive state vector for describing the comprehensive state of each device is formed.

[0048] Specifically, when the user wants to set a scene mode according to the home environment at the first moment, the "Generate Scene" is triggered through the mini program, APP, the screen of the central control panel, voice interaction, etc. At this time, the device status information of each device in the home environment at the first moment is obtained to form a comprehensive state vector for describing the comprehensive status of each device.

[0049] In some embodiments, the step S11 includes:

[0050] Step S111, encoding the device status information of each device in the home environment at the first moment to obtain the status value of each device in multiple states.

[0051] For example, the home environment at the first moment includes an air conditioner and a fan, and the device status of the air conditioner and the device status of the fan at the first moment are as follows:

[0052] Air conditioning: on, cooling mode, 16 degrees, medium speed.

[0053] Fan: On, high speed.

[0054] The coded equipment status of the air conditioner is: 1, 0, 16, 1.

[0055] The device status of the encoded fan is: 1, 2.

[0056] Step S112, writing the state values ​​of each device under the multiple states into a preset data model to generate a device state matrix, wherein each row or each column in the device state matrix represents a device state vector of a device.

[0057] The data model is as follows Figure 2 As shown, the vertical direction represents various devices, such as air conditioners, fans, dehumidifiers, etc., and the horizontal direction represents various states of the devices, such as switches, modes, wind speed, etc.

[0058] Here, the aforementioned home environment at the first moment includes an air conditioner and a fan as an example. The air conditioner state vector in the device state matrix is ​​I1 = [1, 0, 16, 1], and the fan state vector in the device state matrix is ​​I2 = [1, 2].

[0059] Step S113, sequentially concatenate the device state vectors in the device state matrix head to tail to obtain the comprehensive state vector.

[0060] Still taking the aforementioned first moment home environment including air conditioner and fan as an example, the air conditioner state vector I1=[1,0,16,1] and the fan state vector I2=[1,2] in the device state matrix are sequentially concatenated end to end to obtain the comprehensive state vector X=[1,0,16,1,1,2,...].

[0061] Step S12: performing scene classification on the comprehensive state vector using a pre-trained scene classifier to obtain the scene type to which the home environment at the first moment belongs.

[0062] Specifically, the scene classifier can be obtained by training a multi-layer neural network, such as a softmax network. Taking the scene classifier obtained by training a softmax network as an example, the multi-layer neural network structure used by the scene classifier in this embodiment is as follows: Figure 3 The calculation formula of the scene classifier is:

[0063]

[0064] Among them, y i is the calculated classification probability value of the i-th scene type; h i is the i-th scalar element in the vector input to the scene classifier, It's the h of ei Power operation.

[0065] In some embodiments, the scene classifier includes a plurality of preset output scene types. The step S12 comprises: classifying the comprehensive state vector using a pre-trained scene classifier to obtain a classification probability value of each preset scene type to which the home environment belongs at the first moment; and selecting the scene type with the largest probability value from each of the classification probability values ​​as the scene type to which the home environment belongs at the first moment.

[0066] Still taking the aforementioned first moment home environment including air conditioning and fans as an example. The scene classifier includes the preset output scene type. The comprehensive state vector X = [1, 0, 16, 1, 1, 2, ...] is classified using the scene classifier to obtain y1 = 0.0.5, y2 = 0.65, .... Where y2 ≥ y i , i∈[1,n], y i is the classification probability value of the i-th scene type. That is, the classification probability value of y2 is the largest, y2 is the "return home scene", and the scene type to which the home environment belongs at the first moment is the "return home mode".

[0067] In some embodiments, the comprehensive state vector usually contains a large number of low-level features, which may not be sufficient to describe the high-level characteristics of the data. Therefore, in order to capture the high-dimensional features in the comprehensive state vector to more accurately describe the high-level features of the data, the step S12 includes: extracting features from the comprehensive state vector to obtain a high-dimensional feature vector of the comprehensive device state vector; inputting the high-dimensional feature vector into a scene classifier to perform scene classification, and obtaining the scene type to which the home environment belongs at the first moment.

[0068] In one example, feature extraction is performed on the comprehensive device state vector through a deep learning network to obtain the high-dimensional feature vector.

[0069] Step S13, storing the device status information of each device in the home environment at the first moment in association with the scene name of the scene type to which the home environment belongs at the first moment, so as to generate a scene mode corresponding to the scene type.

[0070] This embodiment forms a comprehensive state vector for describing the comprehensive state of each device based on the device state information of each device in the home environment at the first moment; then, the comprehensive state vector is subjected to scene classification using a pre-trained scene classifier to obtain the scene type to which the home environment at the first moment belongs; finally, the device state information of each device in the home environment at the first moment is associated and stored with the scene name of the scene type to which the home environment at the first moment belongs, so as to generate a scene mode corresponding to the scene type. Thus, the scene mode for reproducing the home environment at the first moment is automatically generated.

[0071] Embodiment 2:

[0072] The smart home scene setting method of this embodiment can be applied to electronic devices with communication, computing and data storage capabilities. The specific process can be as follows: Figure 4 As shown, including:

[0073] Step S21: Based on a given scene name, determine a target scene mode corresponding to the scene name from a plurality of preset scene modes. The plurality of preset scene modes are generated by the smart home scene generation method as described in the first embodiment.

[0074] Step S22: extracting target device status information of each device included in the target scene mode.

[0075] Step S23, controlling the device status of each device in the current home environment to be consistent with the target device status information.

[0076] For example, when a user returns home from get off work and wants to set a home mode, the user can provide the scene name they want to set as "home mode" through any interactive method such as a mini program, an APP, the screen of the central control panel, or voice interaction. According to the scene name "home mode" given by the user, the target scene mode corresponding to the scene name "home mode" is determined, and the target device status information of each device included in the target scene mode is extracted. Therefore, according to the extracted target device status information of each device included in the target scene mode, the device status of each device in the current home environment is controlled to be consistent with the target device status information.

[0077] Embodiment three:

[0078] Another embodiment of the present application relates to a smart home scene generation device. The implementation details of the smart home scene generation device of this embodiment are specifically described below. The following content is only for the convenience of understanding the implementation details provided, and is not necessary for the implementation of this solution. The schematic diagram of the smart home scene generation device of this embodiment can be as follows Figure 5As shown, it includes a processing module 301, a scene classification module 302 and a scene generation module 303.

[0079] The processing module 301 is used to form a comprehensive state vector for describing the comprehensive state of each device based on the device state information of each device in the home environment at the first moment acquired.

[0080] Specifically, when the user wants to set a scene mode according to the home environment at the first moment, the "Generate Scene" is triggered through the mini program, APP, the screen of the central control panel, voice interaction, etc. At this time, the device status information of each device in the home environment at the first moment is obtained to form a comprehensive state vector for describing the comprehensive status of each device.

[0081] In some embodiments, the processing module 301 includes an encoding unit, a generating unit and a splicing unit.

[0082] The encoding unit is used to encode the device status information of each device in the home environment at the first moment to obtain the status value of each device in multiple states.

[0083] For example, the home environment at the first moment includes an air conditioner and a fan, and the device status of the air conditioner and the device status of the fan at the first moment are as follows:

[0084] Air conditioning: on, cooling mode, 16 degrees, medium speed.

[0085] Fan: On, high speed.

[0086] The coded equipment status of the air conditioner is: 1, 0, 16, 1.

[0087] The device status of the encoded fan is: 1, 2.

[0088] The generating unit is used to write the state values ​​of the multiple states of each device into a preset data model to generate a device state matrix, wherein each row or each column in the device state matrix represents a device state vector of a device.

[0089] The data model is as follows Figure 2 As shown, the vertical direction represents various devices, such as air conditioners, fans, dehumidifiers, etc., and the horizontal direction represents various states of the devices, such as switches, modes, wind speed, etc.

[0090] Here, the aforementioned home environment at the first moment includes an air conditioner and a fan as an example. The air conditioner state vector in the device state matrix is ​​I1 = [1, 0, 16, 1], and the fan state vector in the device state matrix is ​​O2 = [1, 2].

[0091] The splicing unit is used to sequentially splice the device state vectors in the device state matrix head to tail to obtain the comprehensive state vector.

[0092] Still taking the aforementioned first moment home environment including air conditioner and fan as an example, the air conditioner state vector I1=[1,0,16,1] and the fan state vector I2=[1,2] in the device state matrix are sequentially concatenated end to end to obtain the comprehensive state vector X=[1,0,16,1,1,2,...].

[0093] The scene classification module 302 is used to perform scene classification on the comprehensive state vector using a pre-trained scene classifier to obtain the scene type to which the home environment belongs at the first moment.

[0094] Specifically, the scene classifier can be obtained by training a multi-layer neural network, such as a softmax network.

[0095] In some embodiments, the scene classifier includes a plurality of preset output scene types. The scene classification module 302 includes a classification unit and a selection unit.

[0096] The classification unit is used to classify the comprehensive state vector using a pre-trained scene classifier to obtain a classification probability value of each preset scene type to which the home environment belongs at the first moment.

[0097] The selection unit is used to select the scene type with the largest probability value from each of the classification probability values ​​as the scene type to which the home environment belongs at the first moment.

[0098] Still taking the aforementioned first moment home environment including air conditioning and fans as an example. The scene classifier includes the preset output scene type. The comprehensive state vector X = [1, 0, 16, 1, 1, 2, ...] is classified using the scene classifier to obtain y1 = 0.0.5, y2 = 0.65, .... Where y2 ≥ y i , i∈[1,n], y i is the classification probability value of the i-th scene type. That is, the classification probability value of y2 is the largest, y2 is the "return home scene", and the scene type to which the home environment belongs at the first moment is the "return home mode".

[0099] In some embodiments, the comprehensive state vector usually contains a large number of low-level features, which may not be sufficient to describe the high-level characteristics of the data. Therefore, in order to capture the high-dimensional features in the comprehensive state vector, the high-level features of the data can be described more accurately. The scene classification module 302 includes a feature extraction unit and a classification extraction unit.

[0100] The feature extraction unit is used to extract features from the comprehensive state vector to obtain a high-dimensional feature vector of the comprehensive device state vector.

[0101] In one example, feature extraction is performed on the comprehensive device state vector through a deep learning network to obtain the high-dimensional feature vector.

[0102] The classification extraction unit is used to input the high-dimensional feature vector into a scene classifier to perform scene classification, so as to obtain the scene type to which the home environment belongs at the first moment.

[0103] Specifically, the classification and extraction unit actually includes the above-mentioned classification unit and extraction unit.

[0104] The scene generation module 303 is used to associate and store the device status information of each device in the home environment at the first moment with the scene name of the scene type to which the home environment belongs at the first moment, so as to generate a scene mode corresponding to the scene type.

[0105] In this embodiment, the processing module 301 forms a comprehensive state vector for describing the comprehensive state of each device based on the device state information of each device in the home environment at the first moment; then the scene classification module 302 classifies the comprehensive state vector using a pre-trained scene classifier to obtain the scene type to which the home environment at the first moment belongs; finally, the scene generation module 303 associates and stores the device state information of each device in the home environment at the first moment with the scene name of the scene type to which the home environment at the first moment belongs, so as to generate a scene mode corresponding to the scene type. Thus, the scene mode for reproducing the home environment at the first moment is automatically generated.

[0106] It is worth mentioning that all modules involved in this embodiment are logic modules. In practical applications, a logic unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. In addition, in order to highlight the innovative part of this application, this embodiment does not introduce units that are not closely related to solving the technical problems proposed by this application, but this does not mean that there are no other units in this embodiment.

[0107] Embodiment 4:

[0108] Another embodiment of the present application relates to a smart home scene setting device. The implementation details of a smart home scene setting device of this embodiment are specifically described below. The following content is only for the convenience of understanding the implementation details provided, and is not necessary for the implementation of this solution. The schematic diagram of a smart home scene setting device of this embodiment can be as follows Figure 6As shown, it includes an extraction module 401 and a control module 402.

[0109] The extraction module 401 is used to determine, based on a given scene name, a target scene mode corresponding to the scene name from a plurality of preset scene modes, and extract the target device state information of each device included in the target scene mode; the plurality of preset scene modes are generated by the smart home scene generation method as described in the first embodiment;

[0110] The control module 402 is used to control the device status of each device in the current home environment to be consistent with the target device status information.

[0111] For example, when a user returns home from get off work and wants to set a home mode, the user can provide the scene name he wants to set as "home mode" through any interactive method such as a mini program, an APP, the screen of the central control panel, voice interaction, etc. Then the extraction module 401 determines the target scene mode corresponding to the scene name "home mode" according to the scene name "home mode" given by the user, and extracts the target device status information of each device contained in the target scene mode. Thus, the control module 402 controls the device status of each device in the current home environment to be consistent with the target device status information according to the extracted target device status information of each device contained in the target scene mode.

[0112] It is worth mentioning that all modules involved in this embodiment are logic modules. In practical applications, a logic unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. In addition, in order to highlight the innovative part of this application, this embodiment does not introduce units that are not closely related to solving the technical problems proposed by this application, but this does not mean that there are no other units in this embodiment.

[0113] Embodiment five:

[0114] Another embodiment of the present application relates to an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the smart home scene generation method in the above-mentioned embodiments, or the smart home scene setting method in the above-mentioned embodiments.

[0115] Among them, the memory and the processor are connected in a bus manner, and the bus may include any number of interconnected buses and bridges, and the bus connects various circuits of one or more processors and memories together. The bus can also connect various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be one element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices on a transmission medium. The data processed by the processor is transmitted on a wireless medium via an antenna, and further, the antenna also receives data and transmits the data to the processor.

[0116] The processor is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory can be used to store data used by the processor when performing operations.

[0117] Embodiment six:

[0118] Another embodiment of the present application relates to a computer-readable storage medium storing a computer program, which implements the above method embodiment when executed by a processor.

[0119] That is, those skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, and the program is stored in a storage medium, including a number of instructions to enable a device (which can be a single-chip microcomputer, chip, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as: ROM), random access memory (Random Access Memory, referred to as: RAM), disk or optical disk and other media that can store program codes.

[0120] Those skilled in the art will appreciate that the above embodiments are specific embodiments for implementing the present application, and in actual applications, various changes may be made thereto in form and detail without departing from the spirit and scope of the present application.

Claims

1. A method for generating a smart home scene, characterized in that: include: Based on the acquired device state information of each device in the home environment at the first moment, a comprehensive state vector is formed for describing the comprehensive state of each device; Using a pre-trained scene classifier to perform scene classification on the comprehensive state vector to obtain the scene type to which the home environment at the first moment belongs; The device status information of each device in the home environment at the first moment is associated with the scene name of the scene type to which the home environment at the first moment belongs and stored to generate a scene mode corresponding to the scene type.

2. The method for generating a smart home scene according to claim 1, characterized in that: The scene classifier includes a plurality of preset output scene types; The step of performing scene classification on the comprehensive state vector using a pre-trained scene classifier to obtain the scene type to which the home environment at the first moment belongs includes: Classifying the comprehensive state vector using a pre-trained scene classifier to obtain a classification probability value of each preset scene type to which the home environment belongs at the first moment; The scene type with the largest probability value is selected from each of the classification probability values ​​as the scene type to which the home environment at the first moment belongs.

3. The method for generating a smart home scene according to claim 1, characterized in that: The forming of a comprehensive state vector for describing the comprehensive state of each device based on the acquired device state information of each device in the home environment at the first moment includes: Encode the device status information of each device in the home environment at the first moment to obtain a status value of each device in multiple states; Writing the state values ​​of each device under the multiple states into a preset data model to generate a device state matrix, wherein each row or column in the device state matrix represents a device state vector of a device; The device state vectors in the device state matrix are sequentially concatenated head to tail to obtain the comprehensive state vector.

4. The method for generating a smart home scene according to claim 1, characterized in that: The step of performing scene classification on the comprehensive state vector using a pre-trained scene classifier to obtain the scene type to which the home environment at the first moment belongs includes: Performing feature extraction on the comprehensive state vector to obtain a high-dimensional feature vector of the comprehensive device state vector; The high-dimensional feature vector is input into a scene classifier for scene classification to obtain the scene type to which the home environment belongs at the first moment.

5. The method for generating a smart home scene according to claim 4, characterized in that: The step of extracting features from the comprehensive state vector to obtain a high-dimensional feature vector of the comprehensive device state vector includes: The comprehensive device state vector is subjected to feature extraction through a deep learning network to obtain the high-dimensional feature vector.

6. A method for setting a smart home scene, characterized in that: include: Based on a given scene name, a target scene mode corresponding to the scene name is determined from a plurality of preset scene modes, and target device status information of each device included in the target scene mode is extracted; the plurality of preset scene modes are generated by the smart home scene generation method according to any one of claims 1 to 5; Control the device status of each device in the current home environment to be consistent with the target device status information.

7. A smart home scene generation device, characterized in that: include: A processing module, configured to form a comprehensive state vector for describing a comprehensive state of each device based on the acquired device state information of each device in the home environment at the first moment; A scene classification module, used to classify the comprehensive state vector using a pre-trained scene classifier to obtain the scene type to which the home environment belongs at the first moment; The scene generation module is used to associate and store the device status information of each device in the home environment at the first moment with the scene name of the scene type to which the home environment belongs at the first moment, so as to generate a scene mode corresponding to the scene type.

8. A smart home scene setting device, characterized in that: include: An extraction module, configured to determine, based on a given scene name, a target scene mode corresponding to the scene name from a plurality of preset scene modes, and extract target device status information of each device included in the target scene mode; the plurality of preset scene modes are generated by the smart home scene generation method according to any one of claims 1 to 5; The control module is used to control the device status of each device in the current home environment to be consistent with the target device status information.

9. An electronic device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the smart home scene generation method as described in any one of claims 1 to 5, or execute the smart home scene setting method as described in claim 6.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the smart home scene generation method described in any one of claims 1 to 5 is implemented, or the smart home scene setting method described in claim 6 is implemented.

Images