Systems and methods for human-mounted biosensors and processing biosensor information

1 . A computer-implemented method for performing biosensor-based user authentication, at least a portion of the method being performed by a computing device comprising at least one processor, the method comprising:

receiving user-specific biosignal information from electrodes, wherein the user-specific biosignal information includes electromyography (EMG) information and electrooculography (EOG) information describing at least one facial expression;

extracting the EMG information and the EOG information by applying a bandpass filter to the user-specific biosignal information;

extracting user-specific spatial and temporal representations from the EMG information and the EOG information by:

applying the EMG information and the EOG information to a trained biosignal-based multi-input convolutional neural network (CNN); and

applying an output of the CNN to a Long Short-Term Memory to produce the user-specific spatial and temporal representations; and

applying the extracted user-specific spatial and temporal representations to a machine-learning classifier to produce a machine-learning classifier output indicating user authentication, wherein the machine-learning classifier is trained to identify the extracted user-specific spatial and temporal representations as authentic.

2 . The computer-implemented method of claim 1 , further comprising displaying at least one image of the at least one facial expression on a user display that is a constituent part of:

a helmet;

a hat;

a headset;

a virtual reality headset;

an augmented reality headset;

earbuds;

smart glasses;

a wireless device;

a computer monitor; or

a combination thereof.

3 . The computer-implemented method of claim 1 , wherein the user-specific biosignal information is received from:

a helmet;

a hat;

a headset;

a virtual reality headset;

an augmented reality headset;

earbuds;

smart glasses;

a wireless device;

a computer monitor; or

a combination thereof.

4 . The computer-implemented method of claim 1 , wherein the at least one facial expression is a constituent part of a prescribed sequence of pre-defined facial expressions.

5 . The computer-implemented method of claim 1 , further comprising:

providing, via a user-interface device, a user authentication request; and

restricting receipt of the user-specific biosignal information to a pre-determined time period following the providing the user authentication request.

6 . The computer-implemented method of claim 1 , further comprising performing a multi-factor authentication process, wherein the multi-factor authentication process uses the machine-learning classifier output to authenticate a user.

7 . The computer-implemented method of claim 1 , wherein the output of the CNN describes, using three-dimensional morphable model coefficients, a three-dimensional face mesh.

8 . The computer-implemented method of claim 1 , further comprising:

receiving, from a user interface device and in response to a user prompt to perform at least one facial expression, an input describing authentic user-specific biosignal information; and

preparing a machine-learning classifier training dataset from the authentic user-specific biosignal information.

9 . The computer-implemented method of claim 1 , wherein the machine-learning classifier output is in a plurality of machine-learning classifier outputs corresponding to respective EMG information and respective EOG information describing respective facial expressions, the method further comprising applying the plurality of machine-learning classifier outputs to a hard voting algorithm configured to produce a user authentication output based on a majority vote.

10 . A non-transitory computer-readable medium, comprising processor-executable instructions stored thereon configured to cause a processor to perform:

receiving user-specific biosignal information from electrodes, wherein the user-specific biosignal information includes electromyography (EMG) information and electrooculography (EOG) information describing at least one facial expression;

extracting the EMG information and the EOG information by applying a bandpass filter to the user-specific biosignal information;

extracting user-specific spatial and temporal representations from the EMG information and the EOG information by:

applying the EMG information and the EOG information to a trained biosignal-based multi-input convolutional neural network (CNN); and

applying an output of the CNN to a Long Short-Term Memory to produce the user-specific spatial and temporal representations; and

applying the extracted user-specific spatial and temporal representations to a machine-learning classifier to produce a machine-learning classifier output indicating user authentication, wherein the machine-learning classifier is trained to identify the extracted user-specific spatial and temporal representations as authentic.

11 . The non-transitory computer-readable medium of claim 10 , further comprising processor-executable instructions stored thereon configured to cause the processor to control displaying at least one image of the at least one facial expression on a user display that is a constituent part of:

a helmet;

a hat;

a headset;

a virtual reality headset;

an augmented reality headset;

earbuds;

smart glasses;

a wireless device;

a computer monitor; or

a combination thereof.

12 . The non-transitory computer-readable medium of claim 10 , wherein the user-specific biosignal information is received from:

a helmet;

a hat;

a headset;

a virtual reality headset;

an augmented reality headset;

earbuds;

smart glasses;

a wireless device;

a computer monitor; or

a combination thereof.

13 . The non-transitory computer-readable medium of claim 10 , wherein the at least one facial expression is a constituent part of a prescribed sequence of pre-defined facial expressions.

14 . The non-transitory computer-readable medium of claim 10 , further comprising processor-executable instructions stored thereon configured to cause the processor to:

provide, via a user-interface device, a user authentication request; and

restrict receipt of the user-specific biosignal information to a pre-determined time period following the providing the user authentication request.

15 . The non-transitory computer-readable medium of claim 10 , further comprising processor-executable instructions stored thereon configured to cause the processor to perform a multi-factor authentication process, wherein the multi-factor authentication process uses the machine-learning classifier output to authenticate a user.

16 . The non-transitory computer-readable medium of claim 10 , wherein the output of the CNN describes, using three-dimensional morphable model coefficients, a three-dimensional face mesh.

17 . The non-transitory computer-readable medium of claim 10 , further comprising processor-executable instructions stored thereon configured to cause the processor to:

receive, from a user interface device and in response to a user prompt to perform at least one facial expression, an input describing authentic user-specific biosignal information; and

prepare a machine-learning classifier training dataset from the authentic user-specific biosignal information.

18 . The non-transitory computer-readable medium of claim 10 , further comprising processor-executable instructions stored thereon configured to cause the processor to apply the plurality of machine-learning classifier outputs to a hard voting algorithm configured to produce a user authentication output based on a majority vote, wherein the machine-learning classifier output is in a plurality of machine-learning classifier outputs corresponding to respective EMG information and respective EOG information describing respective facial expressions.

19 . An apparatus configured to perform biosensor-based user authentication, comprising:

a processor; and

a memory communicatively coupled to the processor and storing instructions configured to cause the processor to perform a method comprising:

receiving user-specific biosignal information from electrodes, wherein the user-specific biosignal information includes electromyography (EMG) information and electrooculography (EOG) information describing at least one facial expression;

extracting the EMG information and the EOG information by applying a bandpass filter to the user-specific biosignal information;

extracting user-specific spatial and temporal representations from the EMG information and the EOG information by:

applying the EMG information and the EOG information to a trained biosignal-based multi-input convolutional neural network (CNN); and

applying an output of the CNN to a Long Short-Term Memory to produce the user-specific spatial and temporal representations; and

applying the extracted user-specific spatial and temporal representations to a machine-learning classifier to produce a machine-learning classifier output indicating user authentication, wherein the machine-learning classifier is trained to identify the extracted user-specific spatial and temporal representations as authentic.

20 . The apparatus of claim 19 , wherein the memory further stores instructions configured to cause the processor to control displaying at least one image of the at least one facial expression on a user display that is a constituent part of:

a helmet;

a hat;

a headset;

a virtual reality headset;

an augmented reality headset;

earbuds;

smart glasses;

a wireless device;

a computer monitor; or

a combination thereof.

21 . The apparatus of claim 19 , wherein the user-specific biosignal information is received from:

a helmet;

a hat;

a headset;

a virtual reality headset;

an augmented reality headset;

earbuds;

smart glasses;

a wireless device;

a computer monitor; or

a combination thereof.

22 . The apparatus of claim 19 , wherein the at least one facial expression is a constituent part of a prescribed sequence of pre-defined facial expressions.

23 . The apparatus of claim 19 , wherein the memory further stores instructions configured to cause the processor to:

provide, via a user-interface device, a user authentication request; and

restrict receipt of the user-specific biosignal information to a pre-determined time period following the providing the user authentication request.

24 . The apparatus of claim 19 , wherein the memory further stores instructions configured to cause the processor to perform a multi-factor authentication process, wherein the multi-factor authentication process uses the machine-learning classifier output to authenticate a user.

25 . The apparatus of claim 19 , wherein the output of the CNN describes, using three-dimensional morphable model coefficients, a three-dimensional face mesh.

26 . The apparatus of claim 19 , wherein the memory further stores instructions configured to cause the processor to:

receive, from a user interface device and in response to a user prompt to perform at least one facial expression, an input describing authentic user-specific biosignal information; and

prepare a machine-learning classifier training dataset from the authentic user-specific biosignal information.

27 . The apparatus of claim 19 , wherein the memory further stores instructions configured to cause the processor to apply the plurality of machine-learning classifier outputs to a hard voting algorithm configured to produce a user authentication output based on a majority vote, wherein the machine-learning classifier output is in a plurality of machine-learning classifier outputs corresponding to respective EMG information and respective EOG information describing respective facial expressions.