CN112445288A

CN112445288A - AI semantic recognition device based on electroencephalogram signals

Info

Publication number: CN112445288A
Application number: CN202011128641.7A
Authority: CN
Inventors: 邱和松; 詹丽燕
Original assignee: Individual
Current assignee: Individual
Priority date: 2020-10-21
Filing date: 2020-10-21
Publication date: 2021-03-05

Abstract

The invention belongs to the technical field of semantic recognition, and particularly relates to an electroencephalogram signal AI (artificial intelligence) based semantic recognition device which comprises two signal collectors, an AI semantic common recognizer and wearable display equipment, wherein the two signal collectors are arranged on the head of a human body, the AI semantic common recognizer is used for carrying out semantic recognition on brain waves, and the wearable display equipment is worn on the human body; this device is through setting up the signal collection station of two sucking disc formulas, simultaneously with AI semantic public identifier with dress display device cooperation, it carries out the feature extraction to have through portable equipment of wearing to replace, it is categorised, the semantic database is transferred, semantic identification, the screen result show, the characteristics of simulation vocal production, can adapt to the brain and normally and each function loses old person or the especially language function of serious dyskinesia crowd loses the crowd, utilize AI semantic identification to its wish, the meaning that will want to express is expressed through display screen or loudspeaker, in order to solve the function that this part crowd loses, and this device portability is good, and convenient to use.

Description

AI semantic recognition device based on electroencephalogram signals

Technical Field

The invention belongs to the technical field of semantic recognition, and particularly relates to a semantic recognition device based on electroencephalogram (EEG) signals AI.

Background

In recent years, AI identification is developed and used in a large number of fields, brain electrical identification and body movement control are the highest points for studying conscious morphology and functional movement, how people with normal brain and various disabled elderly or severely disabled people, especially people with language disability, communicate with the brain electrical identification and the like, and how to make sounds and produce actions are the directions which are always pursued by brain researchers in the world at present, namely, the intention of a reader is transmitted to a machine in a certain form of command, so that the realization of an invasive brain-machine interface technology in pigs, monkeys and paralytic patients is successfully proved, and particularly, wireless transmission is realized by a brain-machine chip technology of Mask.

The brain-brain.

Disclosure of Invention

To solve the problems set forth in the background art described above. The invention provides an electroencephalogram signal AI-based semantic recognition device which has the characteristics of feature extraction, classification, semantic database retrieval, semantic recognition, screen result display and simulated sounding through portable substitute equipment.

In order to achieve the purpose, the invention provides the following technical scheme: based on electroencephalogram signal AI semantic recognition device, including installing two signal collector at human head, the AI semantic public recognizer that is used for carrying out semantic recognition with the brain wave and be used for wearing the display device on the human body, the signal collector quantity is two, two the signal collector is installed respectively in the brucard district after the human head left side forehead returns and is located the Wilnik district that the human head left side temporalis wen upward-return, two the signal collector all with AI semantic public recognizer carries out data transmission through wireless signal, AI semantic public recognizer will the signal that the signal collector gathered is received through wireless internet after, to the characteristic extraction of signal degree of depth discernment, categorised, semantic database is transferred, the data after the semantic discernment are spread for dress display device.

Preferably, the signal collector includes sucker, mount pad, signal of telecommunication collector and button cell, the mount pad is fixed in the top intermediate position department of sucker, signal of telecommunication collector and button cell are all installed the inside of mount pad, signal of telecommunication collector with button cell electric connection, signal of telecommunication collector is used for gathering the brain wave signal to give signal transmission dress display device.

Preferably, an adhesive film is arranged on the side wall of the open face of the sucker body, the adhesive film is fixedly adhered to the signal collector, and a sucker cavity for providing a negative pressure space is formed on the sucker body.

Preferably, the top of the mounting seat is hinged with a sealing cover, and the sealing cover is sealed and covered on the top of the mounting seat.

Preferably, the AI semantic common identifier includes a CPU processor, a GPU processor, a memory, and a semantic library unit, the semantic library unit is configured to convert received signal data and perform AI semantic identification, the CPU processor and the GPU processor are configured to perform deep-recognition feature extraction, classification, semantic database retrieval, and semantic identification data, and transmit the data to the wearable display device, and the memory is configured to store the received signal and continuously perfect the semantic library unit.

Preferably, the CPU processor is a 32-core dual-path processor, the video memory capacity of the GPU processor is at least 48G, the memory capacity is at least 40T, and the AI semantic common identifier employs a linux operating system.

Preferably, the wearable display device comprises a display screen, a wristwatch disk body and a wristwatch band, wherein two ends of the wristwatch band are respectively hinged and fixed with two ends of the wristwatch disk body, and the display screen is fixedly mounted on the wristwatch disk body.

Preferably, the inside of watch disk body is equipped with the battery jar that is used for holding the battery, display screen and internal power electric connection.

Preferably, a loudspeaker is arranged on the body of the wristwatch and electrically connected with an internal power supply.

Preferably, a signal processor and a signal receiver are arranged inside the wristwatch disc body, the signal receiver is used for receiving signals transmitted by the AI semantic common identifier, and the signal processor is used for processing the signals and expressing the signals in audio-visual information through the display screen and the loudspeaker.

Compared with the prior art, the invention has the beneficial effects that:

according to the device, two sucker type signal collectors are arranged and are matched with an AI semantic public recognizer and a wearable display device, the device has the characteristics of performing feature extraction, classification, semantic database retrieval, semantic recognition, screen result display and simulated sound production through portable wearing devices, can adapt to the people with normal brain and various loss functions, such as the elderly or people with serious dyskinesia, particularly people with language loss, utilizes AI semantic recognition according to the intention, and expresses the intention to be expressed through a display screen or a loudspeaker, so that the lost functions of the people are solved, and the device is good in portability and convenient to use.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:

FIG. 1 is a schematic structural view of the present invention;

FIG. 2 is a system diagram of an AI semantic common identifier in accordance with the present invention;

FIG. 3 is a system diagram of a wearable display device according to the present invention;

FIG. 4 is a schematic structural diagram of a wearable display device according to the present invention;

fig. 5 is a schematic structural diagram of a signal collector in the present invention.

In the figure: 1. a signal collector; 101. a sucker cavity; 11. a suction cup body; 12. an adhesive film; 13. a mounting seat; 14. an electric signal collector; 15. a button cell; 16. a sealing cover; 2. an AI semantic public identifier; 21. a CPU processor; 22. a GPU processor; 23. a memory; 24. a semantic library unit; 3. a wearable display device; 31. a signal processor; 32. a signal receiver; 33. a display screen; 34. a battery case; 35. a horn; 36. a body of the wristwatch case; 37. a wrist watch band.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Referring to fig. 1-5, the present invention provides the following technical solutions: based on electroencephalogram signal AI semantic recognition device, including installing two signal collector 1 at human head, be used for carrying out AI semantic public recognizer 2 of semantic recognition and be used for wearing display device 3 on the human body with the brain wave, signal collector 1 quantity is two, two signal collector 1 install respectively at the brucard district after the human head left side forehead returns and lie in the welnek district that the human head left side temporal lobe goes up back, two signal collector 1 all carry out data transmission through wireless signal with AI semantic public recognizer 2, AI semantic public recognizer 2 passes through the wireless internet with the signal that signal collector 1 gathered after receiving, to the characteristic extraction of signal depth recognition, categorised, semantic database retrieval, the data is spread for wearing display device 3 after the semantic recognition.

In this embodiment, two signal collectors 1 are respectively arranged in a bruck area and a Wilnik area of a brain in a manner of combining a flat suction cup body 11 with an adhesive film 12, the signal collectors 14 are used for signal collection, amplification and pretreatment, then signals are transmitted to a portable wearable display device 3 through a radio transmitting device, feature extraction, classification, semantic database retrieval, semantic identification, screen result display and simulated sound production are performed through the portable wearable display device 3, wherein the signal collectors 1 are horn-type flat wide-mouth suction cup bodies 11 with the height of 3mm, the suction cup bodies 11 are internally provided with the signal collectors 14, the inner walls of the suction cup bodies 11 are covered with disposable efficient breathable waterproof adhesive films 12, mounting seats 13 on the suction cup bodies 11 are internally provided with replaceable button batteries 15, and the signal collectors 14 are used for signal collection, Amplifying and preprocessing are transmitted to an AI semantic common recognizer 2 through a wireless internet, wherein the AI semantic common recognizer 2 (a background-arranged super computer) consists of a CPU32 core double-path thread, a GPU48G, a storage 40T, a continuously perfect semantic library unit 24 and a linux operating system, signals are received through the wireless internet and then enter the AI semantic common recognizer 2 for data conversion, deep recognition feature extraction, classification, semantic database calling and data transmission after semantic recognition are realized to a wearable display device 3, the wearable display device 3 displays semantics on a display screen 33 according to signals and instructions sent by the AI semantic common recognizer 2 and sends out sounds through a loudspeaker 35 after semantic recognition, the device is matched with the AI semantic common recognizer 2 and the wearable display device 3 through arranging two sucker type signal collectors 1, can adapt to the normal and each functional disability old person of brain or serious dyskinesia crowd especially language function losing crowd, utilize AI semantic recognition to its will, express the meaning that wants to express through display screen 33 or loudspeaker 35 to solve the function that this part crowd loses, and this device portability is good, facilitates the use.

Specifically, the signal collector 1 comprises a suction cup body 11, a mounting seat 13, an electric signal collector 14 and a button battery 15, the mounting seat 13 is fixed at the middle position of the top of the suction cup body 11, the electric signal collector 14 and the button battery 15 are both installed inside the mounting seat 13, the electric signal collector 14 is electrically connected with the button battery 15, the electric signal collector 14 is used for collecting brain wave signals and transmitting the signals to the wearable display device 3, an adhesive film 12 is arranged on the side wall of the open face of the suction cup body 11, the adhesive film 12 is fixedly adhered to the signal collector 1, a suction cup cavity 101 for providing a negative pressure space is formed on the suction cup body 11, a sealing cover 16 is hinged to the top of the mounting seat 13, the sealing cover 16 is hermetically covered on the top of the mounting seat 13, wherein the signal collector 1 is a horn-type wide mouth suction cup body 11 with a height of 3mm, the electric, the inner wall of the sucker body 11 is covered with a disposable high-efficiency breathable waterproof adhesive film 12, a replaceable button battery 15 is arranged in a mounting seat 13 on the sucker body 11, and an electric signal collector 14 is used for realizing signal acquisition, amplification and pretreatment and transmitting the signals to the AI semantic public identifier 2 through a wireless internet.

Specifically, the AI semantic common identifier 2 includes a CPU processor 21, a GPU processor 22, a memory 23 and a semantic library unit 24, the semantic library unit 24 is configured to convert received signal data to perform AI semantic identification, the CPU processor 21 and the GPU processor 22 are configured to perform deep identification, feature extraction, classification, semantic database retrieval, semantic identification, and transmit data to the wearable display device 3, the memory 23 is configured to store received signals and continuously perfect the semantic library unit 24, the CPU processor 21 is a 32-core dual-path line, the display memory capacity of the GPU processor 22 is at least 48G, the capacity of the memory 23 is at least 40T, the AI semantic common identifier 2 employs a linux operating system, the semantic AI common identifier 2 is configured by a CPU32 core dual-path thread, a GPU48G, a memory 40T, a continuously perfect semantic library unit 24, and a linux operating system, signals are received through a wireless internet and then enter the semantic AI common identifier 2 to perform data conversion, and the deep recognition features extraction, classification, semantic database retrieval and semantic recognition data are transmitted to the wearable display device 3.

Specifically, the wearable display device 3 comprises a display screen 33, a wristwatch disk 36 and a wristwatch strap 37, wherein two ends of the wristwatch strap 37 are respectively hinged and fixed with two ends of the wristwatch disk 36, the display screen 33 is fixedly installed on the wristwatch disk 36, a battery jar 34 for accommodating a battery is arranged inside the wristwatch disk 36, the display screen 33 is electrically connected with an internal power supply, a loudspeaker 35 is arranged on the wristwatch disk 36, the loudspeaker 35 is electrically connected with the internal power supply, a signal processor 31 and a signal receiver 32 are arranged inside the wristwatch disk 36, the signal receiver 32 is used for receiving signals transmitted by the AI semantic public identifier 2, the signal processor 31 is used for processing the signals, the wearable display device 3 displays the semantics on the display screen 33 according to the signal and the instruction sent by the AI semantic public recognizer 2, and sends the sound through the loudspeaker 35 after the recognition according to the semantics.

The working principle and the using process of the invention are as follows: the two signal collectors 1 are respectively arranged in a brucard area and a Wilnik area of a brain in a mode of combining a flat suction cup body 11 with an adhesive film 12, the signal collectors 14 are used for realizing signal collection, amplification and pretreatment, then signals are transmitted to a portable wearable display device 3 through a radio transmitting device, feature extraction, classification, semantic database calling, semantic recognition, screen result display and simulated sound production are carried out through the portable wearable display device 3, wherein the signal collectors 1 are horn-shaped flat wide-mouth suction cup bodies 11 with the height of 3mm, the suction cup bodies 11 are internally provided with the signal collectors 14, the inner walls of the suction cup bodies 11 are covered with disposable high-efficiency breathable waterproof adhesive films 12, replaceable button batteries 15 are arranged in mounting seats 13 on the suction cup bodies 11, and the signal collectors 14 are used for realizing signal collection, amplification and sound production, The preprocessing is transmitted to the AI semantic common recognizer 2 through the wireless internet, wherein the AI semantic common recognizer 2 consists of a CPU32 core double-path thread, a GPU48G, a memory 40T, a continuously improved semantic library unit 24 and a linux operating system, signals are received through the wireless internet and then enter the AI semantic common recognizer 2 for data conversion, deep recognition feature extraction, classification, semantic database retrieval and data transmission after semantic recognition to the wearable display device 3 are realized, the wearable display device 3 displays semantics on a display screen 33 according to signals and instructions sent by the AI semantic common recognizer 2, and sounds are sent through a loudspeaker 35 after semantic recognition.

Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that changes may be made in the embodiments and/or equivalents thereof without departing from the spirit and scope of the invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. Based on electroencephalogram signal AI semantic recognition device which characterized in that: comprises two signal collectors (1) arranged on the head of a human body, an AI semantic common recognizer (2) used for carrying out semantic recognition on brain waves and a wearable display device (3) worn on the human body, the number of the signal collectors (1) is two, the two signal collectors (1) are respectively arranged in a bruca area behind the forehead on the left side of the head of a human body and a Weirnike area on the temporal lobe on the left side of the head of the human body, the two signal collectors (1) and the AI semantic common recognizer (2) carry out data transmission through wireless signals, the AI semantic common recognizer (2) receives the signals collected by the signal collector (1) through the wireless internet, and (3) carrying out feature extraction and classification on signal depth recognition, calling a semantic database, and transmitting data after semantic recognition to the wearable display device (3).

2. The electroencephalography signal AI based semantic recognition apparatus according to claim 1, characterized in that: signal collector (1) includes sucker (11), mount pad (13), signal of telecommunication collector (14) and button cell (15), mount pad (13) are fixed in the top intermediate position department of sucker (11), signal of telecommunication collector (14) and button cell (15) are all installed the inside of mount pad (13), signal of telecommunication collector (14) with button cell (15) electric connection, signal of telecommunication collector (14) are used for gathering the brain wave signal to give signal transmission dress display device (3).

3. The electroencephalography signal AI based semantic recognition apparatus according to claim 2, characterized in that: an adhesive film (12) is arranged on the side wall of the open face of the sucker body (11), the adhesive film (12) is fixedly adhered to the signal collector (1), and a sucker cavity (101) for providing a negative pressure space is formed on the sucker body (11).

4. The electroencephalography signal AI based semantic recognition apparatus according to claim 3, characterized in that: the top of the mounting seat (13) is hinged with a sealing cover (16), and the sealing cover (16) is hermetically covered on the top of the mounting seat (13).

5. The electroencephalography signal AI based semantic recognition apparatus according to claim 1, characterized in that: the AI semantic common recognizer (2) comprises a CPU (central processing unit) processor (21), a GPU (graphics processing unit) processor (22), a memory (23) and a semantic library unit (24), wherein the semantic library unit (24) is used for converting received signal data and recognizing AI semantics, the CPU processor (21) and the GPU processor (22) are used for extracting and classifying deep-recognized features, calling a semantic database and recognizing the semantic data and transmitting the deep-recognized data to the wearable display device (3), and the memory (23) is used for storing the received signals and continuously perfecting the semantic library unit (24).

6. The electroencephalography signal AI-based semantic recognition apparatus of claim 5, wherein: the CPU processor (21) is a 32-core dual-path processor, the video memory capacity of the GPU processor (22) is at least 48G, the capacity of the memory (23) is at least 40T, and the AI semantic common identifier (2) adopts a linux operating system.

7. The electroencephalography signal AI based semantic recognition apparatus according to claim 1, characterized in that: dress display device (3) including display screen (33), watch disk body (36) and watch area (37), the both ends of watch area (37) respectively with the both ends of watch disk body (36) are articulated fixed, display screen (33) fixed mounting be in on the watch disk body (36).

8. The electroencephalography signal AI based semantic recognition apparatus according to claim 7, characterized in that: the inside of watch disk body (36) is equipped with battery jar (34) that is used for holding the battery, display screen (33) and internal power electrical connection.

9. The electroencephalography signal AI based semantic recognition apparatus according to claim 8, characterized in that: the wristwatch disc body (36) is provided with a loudspeaker (35), and the loudspeaker (35) is electrically connected with an internal power supply.

10. The electroencephalography signal AI based semantic recognition apparatus according to claim 9, characterized in that: the wrist watch comprises a wrist watch disc body (36), wherein a signal processor (31) and a signal receiver (32) are arranged inside the wrist watch disc body (36), the signal receiver (32) is used for receiving signals transmitted by the AI semantic common recognizer (2), the signal processor (31) is used for processing the signals, and the signals are expressed in audio-visual information through a display screen (33) and a loudspeaker (35).