Intelligent brain wave control lighting system based on convolutional neural network

Abstract

The utility model provides an intelligent brain wave control lighting system based on a convolutional neural network, which belongs to the technical field of lighting control and comprises a brain wave analysis system, a head-mounted device and a lighting control device, the head-mounted device is connected with the brain wave analysis system, and the lighting control device is connected with the brain wave analysis system. The brain wave analysis system and the light control device are both arranged on the table lamp. The system further comprises a mode selection device, a voice acquisition and recognition device and a playing device, the mode selection device is arranged on the table lamp corresponding to the light control device, and the voice acquisition and recognition device is arranged on the head-mounted device or the light control device. The method can better adapt to personalized requirements of users, the users can control light through brain waves, the method is a more visual and natural interaction mode, the usability is higher, the physiological state of the users is monitored in real time, and the user experience is improved. The intelligent table lamp can dynamically adjust the frequency and brightness of the light source, and the intelligent head-mounted device can play appropriate music according to the physiological state of the user.

Description

Technical Field

[0001] The utility model relates to the field of lighting control, in particular to an intelligent brainwave-controlled lighting system based on a convolutional neural network. Background Art

[0002] In real life, traditional table lamps mainly provide basic lighting functions and are usually placed on desks or furniture to assist users in learning or working. With the development of technology, intelligent table lamps have incorporated more functions, such as timing dimming, light sensing, etc., to adapt to different usage scenarios. Existing intelligent table lamp technologies rely on preset work and rest schedules to adjust the light source. Research shows that different brainwave states (alpha waves, beta waves, theta waves, delta waves) are closely related to a person's activity states (such as learning, working, resting, sleeping). Existing intelligent table lamps attempt to guide the user's brainwave state by adjusting the light source frequency to improve work efficiency or rest quality, but they cannot real-time sense and adapt to the user's actual brainwave state, and there are limitations in personalized adjustment and real-time response. Brainwaves mainly have 4 states and are classified into alpha waves, beta waves, theta waves, and delta waves according to their frequencies. The better states for learning and working are when the brainwaves are in the alpha wave or beta wave state, and the better states for resting and sleeping are when the brainwaves are in the theta wave or delta wave state.

[0003] Chinese Patent with Publication No. CN 219780451 U discloses an intelligent lighting system, which includes: a brainwave acquisition device, a lighting device, and a control device. The lighting device includes a lamp and a light brightness control circuit, and the light brightness control circuit is connected to the lamp. The control device is respectively connected to the brainwave acquisition device and the light brightness control circuit. Among them, the control device is used to receive the data of the brainwave acquisition device and control the light brightness control circuit based on the data, so that the light brightness control circuit adjusts the illumination intensity of the lamp. This application can collect the user's brainwave data through the brainwave acquisition device and receive the data of the brainwave acquisition device through the control device, and can detect the user's physiological state in real time. Then, the control device controls the light brightness control circuit to adjust the illumination intensity of the lamp, and can adjust the brightness of the lamp in real time based on the user's actual physiological state, so as to meet the user's actual needs. The whole system is controlled according to brainwaves throughout the process. However, when the user needs to work overtime or needs to be controlled according to a set mode, the lighting cannot be controlled. For example, when the user needs to work overtime, but the brain is sleepy, at this time the light may only be adjusted to become dim, making the user adapt to rest. However, the user often needs to force an overtime mode and cannot set the required mode, and at the same time, voice control cannot be performed, which will bring troubles to some users. Therefore, this application designs a more intelligent lighting control system. Summary of the Utility Model

[0004] The purpose of the present utility model is to provide an intelligent brainwave-controlled lighting system based on a convolutional neural network, so as to solve the technical problem that the existing intelligent lighting system can only be controlled according to brainwaves and cannot meet the needs of users in different modes.

[0005] Through the initial setting of the user, the lighting is adjusted according to the actual setting of the user. At the same time, according to the set mode and the state data of the brainwaves collected from the person, the lighting brightness is then adjusted to achieve light stimulation of the human brain. For example, in the overtime mode set by the user, when the brainwaves indicating drowsiness appear in the human brain, the light of the lamp can be controlled to be brighter to stimulate the human brain, making the person become energetic from drowsiness, and realizing the coordination of the specific mode and brainwaves.

[0006] In order to achieve the above purpose, the technical solution adopted by the present utility model is as follows:

[0007] An intelligent brainwave-controlled lighting system based on a convolutional neural network includes a brainwave analysis system, a head-mounted device, and a lighting control device. The head-mounted device is connected to the brainwave analysis system, and the lighting control device is connected to the brainwave analysis system. Both the brainwave analysis system and the lighting control device are provided on a table lamp. It also includes a mode selection device, a voice collection and recognition device, and a playback device. The mode selection device is provided on the table lamp corresponding to the lighting control device. The voice collection and recognition device is provided on the head-mounted device or the lighting control device. The playback device is provided on the head-mounted device or the table lamp corresponding to the lighting control device.

[0008] Further, the mode selection device is a rotary switch, and the rotary switch is connected to the lighting control device. When one of the buttons on the rotary switch is pressed, the lighting control device controls the lighting according to the mode set by the pressed mode button. The mode selection device adopts a three-stage knob design, and the three-stage knob is directly connected to the lighting control device. When the user rotates and positions to a specific level of the knob, the lighting control device will control the lighting according to the mode represented by this level.

[0009] Further, the head-mounted device includes a U-shaped head-mounted body, in-ear headphones, a music player, a communication and storage circuit case, an emotion perception lamp, and a brainwave detection sensor. The brainwave detection sensor is provided on the inner side of the U-shaped head-mounted body. The communication and storage circuit case is provided on the outer side of the U-shaped head-mounted body. The communication and storage circuit case is provided with a brainwave transmission communication module and a music storage module. The music player is provided on the side of the communication and storage circuit case. The in-ear headphones are connected to the music player through a wire. The emotion perception lamp is provided on the top of the music player. The emotion perception lamp is connected to the brainwave transmission communication module. The brainwave transmission communication module transmits the collected brainwave data to the brainwave analysis system.

[0010] Further, the electroencephalogram analysis system includes an NVIDIA control panel and a wireless communication module. The wireless communication module is connected to the NVIDIA control panel, and a pre-trained Mobilenet neural network model is loaded in the NVIDIA control panel. After the NVIDIA control panel processes the electroencephalogram data, it transmits the specific on-demand type to the lighting control device, and the lighting control device controls the lighting mode.

[0011] Further, the voice acquisition and recognition device includes a microphone, a voice recognition circuit, and a voice storage chip. The microphone collects the voice and transmits it to the voice recognition circuit. The voice recognition circuit transmits the recognized voice to the voice storage chip for comparison, and then transmits the comparison result to the lighting control device.

[0012] Further, the playback device includes a music player and an audio player. The music player is connected to the electroencephalogram analysis system, and the music player downloads or stores music in advance.

[0013] Due to the adoption of the above technical solutions, the present utility model has the following beneficial effects:

[0014] The present utility model can better meet the personalized needs of users. Users can control the lighting through their thoughts, providing a more intuitive and natural interaction method without the need to manipulate physical devices. For some special groups, such as the disabled, it has higher usability. According to the real-time monitored physiological state of the user, the intelligent desk lamp can dynamically adjust the frequency and brightness of the light source, and the intelligent head-mounted device can play appropriate music according to the user's physiological state. For example, when it is monitored that the user is in a highly concentrated working state, increase the brightness and frequency of the light source and play music with a faster rhythm to enhance alertness and attention. Analyze the user's usage habits and physiological state data to provide more personalized and accurate services. Through the analysis of the user's historical data, the intelligent desk lamp can continuously optimize the adjustment strategy to better meet the needs of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is the system block diagram of Embodiment 1 of the present utility model;

[0016] Figure 2 is the system block diagram of Embodiment 2 of the present utility model;

[0017] Figure 3 is the front three-dimensional view of the head-mounted device of the present utility model;

[0018] Figure 4 is the side three-dimensional view of the head-mounted device of the present utility model;

[0019] Figure 5 is the front view of the head-mounted device of the present utility model;

[0020] Figure 6 It is the left view of the head-mounted device of the present utility model;

[0021] Figure 7 It is the left view flipping diagram of the head-mounted device of the present utility model;

[0022] Figure 8 It is the schematic structural diagram of the lamp body of the present utility model;

[0023] Figure 9 It is the schematic side structure diagram of the lamp body of the present utility model.

[0024] Reference numerals in the figure: 1 - U-shaped head-mounted body; 2 - in-ear headphones; 3 - music player; 4 - communication and storage circuit housing; 5 - emotion perception lamp; 6 - electroencephalogram detection sensor; 7 - lamp base; 8 - rotary switch; 9 - bendable adjustment rod; 10 - lamp panel. Specific embodiments

[0025] To make the purpose, technical solutions and advantages of the present utility model clearer, the following takes preferred embodiments as examples and further elaborates on the present utility model with reference to the accompanying drawings. However, it should be noted that many details listed in the specification are only for enabling the reader to have a thorough understanding of one or more aspects of the present utility model, and these aspects of the present utility model can be implemented even without these specific details.

[0026] Embodiment 1:

[0027] As Figure 1 shown, the intelligent electroencephalogram-controlled lighting system based on a convolutional neural network includes an electroencephalogram analysis system, a head-mounted device, and a lighting control device. The head-mounted device is connected to the electroencephalogram analysis system, and the lighting control device is connected to the electroencephalogram analysis system. Both the electroencephalogram analysis system and the lighting control device are provided on the table lamp. It also includes a mode selection device, a voice collection and recognition device, and a playback device. The mode selection device is provided on the table lamp corresponding to the lighting control device, the voice collection and recognition device is provided on the head-mounted device, and the playback device is provided on the head-mounted device. Both the voice collection and recognition device and the playback device are provided on the head-mounted device. At this time, voice collection, music playback, or radio playback are all on the head-mounted device. The advantage of this is that there is no need for external speakers, which is equivalent to playing through a pair of headphones and will not cause interference to the external environment.

[0028] The control lighting system is provided on the lamp body. As Figures 8-9 shown, the lamp body includes a lamp base 7, a rotary switch 8, a bendable adjustment rod 9, and a lamp panel 10. The rotary switch 8 is provided on the lamp base 7. One end of the bendable adjustment rod 9 is connected to the lamp base 7, and the other end is connected to the lamp panel 10. Both the electroencephalogram analysis system and the lighting control device are provided inside the lamp base 7.

[0029] The head-mounted functional unit includes music of different playlists. The lighting execution unit includes a controller that outputs a square wave based on a timer and controls the brightness through the duty cycle. The lighting functional unit includes controlling the brightness and warm / cool color of the light. The output end of the electroencephalogram detection sensor is connected to the input end of the intelligent data analysis system. The output end of the intelligent data analysis system is connected to the input end of the lighting system controller and the input end of the music player. The output end of the lighting system controller is used to control the brightness and warm / cool color of the light, and the output end of the music player is used to switch to a suitable music playlist. After the electroencephalogram signal is converted into physiological data, the intelligent data analysis system automatically switches to a suitable lighting brightness, warm / cool color, and a suitable music playlist by processing and matching the physiological data.

[0030] The light intensity is divided into three categories: The first category of intensity meets the needs of users for highly efficient and concentrated work, with a light intensity of 25 - 35 degrees; the second category of intensity is suitable for highly efficient and concentrated work, with a light intensity of 35 - 40 degrees, and when the user feels sleepy, the light reminds the user to stay awake; the third category of intensity is suitable for the user's relaxation or rest state, with a light intensity ranging from 5 to the off state, creating a lighting atmosphere suitable for rest.

[0031] When giving a reminder, it can be done by controlling the brightness or by controlling the flicker through the frequencies of alpha waves, beta waves, theta waves, and delta waves to achieve the effect of refreshing or hypnotizing people.

[0032] In this embodiment, the mode selection device is a rotary switch, which is set as a three-stage knob. Then this knob has two modes: up and down. The rotary switch is connected to the lighting control device. When the rotary switch rotates to a specific gear, the lighting control device controls the light according to the mode set by the rotary switch. The working modes of the table lamp are divided into three modes: (When the knob is in the first gear, the voice interaction function is not turned on) The first is to control the light brightness according to brain waves, and the voice interaction function is not turned on. (When the knob is in the pressed state, it is the forced working mode; if the knob is in the unpressed state, it is the free working mode) One is the brain wave forced working and learning mode. At this time, the light intensity is 25-30. When θ waves or δ waves are recognized by the brain wave, it means that the user may be sleepy, but needs to work and study. The light intensity increases to prevent the user from feeling sleepy during the highly concentrated working period and affecting work efficiency. At the same time, the light can also be controlled to flash by the α wave frequency, so as to induce the human brain to generate α waves and achieve the effect of refreshing. The other is the working brain wave free working and learning mode. When θ waves or δ waves are recognized, it means that the user is sleepy, and the brightness is lowered to make the user gradually enter the sleep mode. (When the knob is in the second gear, the voice interaction function is turned on) The second is that the brain wave and voice interaction jointly control the light brightness to achieve the satisfaction of the user (the music playback function is turned off in the first two modes). When θ waves or δ waves appear in the brain wave, it means that the user is sleepy. If there is no voice interaction to control, it will flash or the brightness will be lowered at the frequency of θ waves or δ waves. If there is voice interaction, it will be controlled according to the voice. (When the knob is in the second gear, the music interaction function is turned on) The third is that the brain wave and music interaction jointly control, and at the same time, corresponding music that matches the scene is played according to the current brain wave state of the user to create an atmosphere that the user likes. The music corresponding to the corresponding brain wave is marked when the music data is stored.

[0033] In this embodiment, as Figures 3-7As shown in the figure, the head-mounted device includes a U-shaped head-mounted body 1, in-ear headphones 2, a music player 3, a communication and storage circuit housing 4, an emotion-sensing light 5, and an electroencephalogram (EEG) detection sensor 6. The EEG detection sensor 6 is disposed on the inner side of the U-shaped head-mounted body 1, and the communication and storage circuit housing 4 is disposed on the outer side of the U-shaped head-mounted body 1. The communication and storage circuit housing 4 is provided with an EEG transmission and communication module and a music storage module. The music player 3 is disposed on the side of the communication and storage circuit housing 4. The in-ear headphones 2 are connected to the music player 3 through a wire. The emotion-sensing light 5 is disposed on the top of the music player 3, and the emotion-sensing light 5 is connected to the EEG transmission and communication module. The EEG transmission and communication module transmits the collected EEG data to an EEG analysis system. The head-mounted device includes a U-shaped head-mounted body and an EEG detection sensor. A device containing a head-mounted wireless communication module and a storage unit is disposed on the U-shaped head-mounted body. The device is also provided with an emotion-sensing light so that the user or others can know the brain state of the user. A music player with wired in-ear headphones is also disposed on the head-mounted device. The EEG detection sensor is disposed on the inner side of the U-shaped head-mounted body. The U-shaped head-mounted body is worn on the user's head, and the EEG detection sensor is connected to the head-mounted wireless communication module.

[0034] In this embodiment, the EEG analysis system includes an NVIDIA control panel and a wireless communication module. The wireless communication module is connected to the NVIDIA control panel. A pre-trained Mobilenet neural network model is loaded into the NVIDIA control panel. The Mobilenet neural network model is used to extract high-level features from the EEG data. This deep learning model can more accurately understand the user's EEG pattern and provide more accurate decision-making for intelligent adjustment. The intelligent EEG control system can better adapt to the personalized needs of users. At the same time, the user can control the light through the mind, providing a more intuitive and natural interaction method. There is no need to operate physical devices, which has higher usability for some special groups such as disabled people. The output of the NVIDIA control panel is an alpha wave, a beta wave, a theta wave, or a delta wave. A well-known waveform is output and then given to the control device to adjust the light. The part of the light adjustment structure belongs to the existing structure and will not be described in detail. The optimal states for learning and working are when the EEG is in the alpha wave or beta wave state, and the optimal states for rest and sleep are when the EEG is in the theta wave or delta wave state. The NVIDIA control panel is an existing module that can be purchased in the market. The Mobilenet neural network model is trained through an external computer. The alpha wave, beta wave, theta wave, and delta wave are the detection target information. Then, the trained model is obtained by inputting a number of collected EEG data for training. This process can be achieved on a computer. Then, the trained model is downloaded to the NVIDIA control panel for use. The NVIDIA control panel uses the model number NVIDIAjetsonxavier nx.

[0035] In this embodiment, the voice acquisition and recognition device includes a microphone, a voice recognition circuit, and a voice storage chip. The microphone collects voice and transmits it to the voice recognition circuit. The voice recognition circuit transmits the recognized voice to the voice storage chip for comparison, and then transmits the comparison result to the lighting control device.

[0036] In this embodiment, the playback device includes a music player, an audio player, and a music selection screen. The music selection screen is connected to both the music player and the audio player. The music player is connected to the electroencephalogram analysis system, and the music player downloads or stores music in advance.

[0037] The specific working process is as follows:

[0038] First, use an external computer to train with the Mobilenet neural network model. Input α waves, β waves, θ waves, and δ waves as target information, and then input a number of collected electroencephalogram data for training to obtain a trained model. This process is not the innovative content of this application and is prepared in advance with existing information and belongs to the existing training process. Input the trained model into the NVIDIA control panel. After the head-mounted device collects the electroencephalogram data of the human brain and transmits it to the processing unit, as Figure 1 , it can be wired transmission or wireless, and wireless is convenient. Then the processing unit identifies what specific wave the collected electroencephalogram data is. For example, when it is a δ wave or a θ wave, and the user selects the first type of controlling the brightness of the light to increase when forced to work overtime or study according to the electroencephalogram, or controlling the light to flash according to the frequency of the α wave, making people become energetic. In the free work and study mode, if α waves or β waves are collected, the brightness of the light is increased, and when δ waves or θ waves are collected, the brightness is decreased. If it is the second type, the electroencephalogram and voice interaction jointly control the brightness of the light. When θ waves or δ waves appear in the electroencephalogram, it means there is drowsiness. If there is no voice control, it flashes or the brightness is lowered at the frequency of θ waves or δ waves. If there is voice, it is controlled according to the voice; in the third type, the electroencephalogram and music cooperate to control, and corresponding music that matches the scene is played according to the current electroencephalogram state of the user to create a favorite atmosphere for the user. The music corresponding to the corresponding electroencephalogram is marked when the music data is stored.

[0039] Comparative Example 1:

[0040] This Embodiment 1 is compared with the Chinese patent with the publication number CN 219780451 U for an intelligent lighting system. The existing one is controlled according to the user's brain waves. However, when people need to work or study overtime compulsorily, there is no way to give an active reminder or adjust the light to make people more concentrated at work. But this application can achieve this. Before the user uses it, they can select the working mode of compulsory overtime. At this time, the light can be adjusted to remind people to concentrate on work or study. At the same time, this application also adds voice combined control, which cannot be achieved in the comparative example.

[0041] Embodiment 2

[0042] As Figure 2 shown, the difference between this embodiment and Embodiment 1 is that both the voice collection and recognition device and the playback device are directly set on the table lamp. At this time, the music playback is external, which is more suitable for use in some private rooms, but not suitable in some public places, such as libraries.

[0043] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principle and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

[0044] Matters not covered by the present invention are well-known technologies.

[0045] The above embodiments are only used to illustrate the technical concept and characteristics of the present invention. The purpose is to enable those who are familiar with this technology to understand the content of the present invention and implement it accordingly, and it cannot be used to limit the protection scope of the present invention. Any equivalent changes or modifications made according to the spirit of the present invention should be covered within the protection scope of the present invention.

Claims

1. An intelligent brainwave-controlled lighting system based on a convolutional neural network, comprising a brainwave analysis system, a head-mounted device and a lighting control device, wherein the head-mounted device is connected to the brainwave analysis system, and the lighting control device is connected to the brainwave analysis system, and the brainwave analysis system and the lighting control device are both arranged on a desk lamp, characterized in that: It also includes a mode selection device, a voice collection and recognition device, and a playback device. The mode selection device is arranged on a desk lamp corresponding to the light control device, the voice collection and recognition device is arranged on a head-mounted device or on the light control device, and the playback device is arranged on the head-mounted device or on the desk lamp corresponding to the light control device.

2. The intelligent brainwave controlled lighting system based on convolutional neural network according to claim 1, characterized in that: The mode selection device is a rotary switch, and the rotary switches are connected to the light control device. When a button in the rotary switch is pressed, the light control device controls the light according to the mode set by the pressed mode button.

3. The intelligent brainwave controlled lighting system based on convolutional neural network according to claim 1, characterized in that: The head-mounted device comprises a U-shaped head-mounted body (1), an in-ear headset (2), a music player (3), a communication and storage circuit shell (4), an emotion perception lamp (5) and a brain wave detection sensor (6); the brain wave detection sensor (6) is arranged on the inner side of the U-shaped head-mounted body (1); the communication and storage circuit shell (4) is arranged on the outer side of the U-shaped head-mounted body (1); the communication and storage circuit shell (4) is provided with a brain wave transmission communication module and a music storage module; the music player (3) is arranged on the side of the communication and storage circuit shell (4); the in-ear headset (2) and the music player (3) are connected through a wire; the emotion perception lamp (5) is arranged on the top of the music player (3); the emotion perception lamp (5) is connected to the brain wave transmission communication module; and the brain wave transmission communication module transmits the collected brain wave data to a brain wave analysis system.

4. The intelligent brainwave controlled lighting system based on convolutional neural network according to claim 1, characterized in that: The brain wave analysis system includes an NVIDIA control panel and a wireless communication module. The wireless communication module is connected to the NVIDIA control panel, and the NVIDIA control panel is loaded with an existing trained Mobilenet neural network model.

5. The intelligent brainwave controlled lighting system based on convolutional neural network according to claim 1, characterized in that: The voice collection and recognition device includes a microphone, a voice recognition circuit and a voice storage chip. The microphone collects voice and transmits it to the voice recognition circuit. The voice recognition circuit transmits the recognized voice to the voice storage chip for comparison, and then transmits the comparison result to the lighting control device.

6. The intelligent brainwave controlled lighting system based on convolutional neural network according to claim 1, characterized in that: The playing device includes a music player, an audio player and a music selection screen. The music selection screen is connected to the music player and the audio player. The music player is connected to the brain wave analysis system. The music player downloads or stores music in advance.

Images