GESTURE RECOGNITION APPPARATUS AND METHOD

• US 20120106298A1
• Filed: 10/28/2010
• Published: 05/03/2012
• Est. Priority Date: 10/28/2010
• Status: Active Grant

• Current Assignee: Hon Hai Precision Industry Co., Ltd., Massachusetts Institute of Technology
• Original Assignee: Hon Hai Precision Industry Co., Ltd., Massachusetts Institute of Technology
• Inventors: LIU, LING-WEI
• Granted Patent: US 8,223,589 B2
• Time in Patent Office: Days
• US Class Current: 367/87
• CPC Class Codes: G06F 18/2135 based on approximation criteria, e.g. principal component analysis
• CPC Class Codes: G06V 40/20 Movements or behaviour, e.g. gesture recognition recognition of facial expressions G06V40/16

Abstract

The disclosure provides a gesture recognition apparatus and method. The gesture recognition apparatus includes an ultrasound transmitter, an ultrasound receiver, a dividing module, a computing module, a gesture library, and a recognition module. The dividing module is configured to divide reflected ultrasound signals into a plurality of frames according to time intervals. The computing module is configured to obtain an eigenvalue of each frame. The classifying module is configured to filter the eigenvalues to obtain gesture eigenvalues, and to obtain a matrix of probabilities of the gesture eigenvalues. The recognition module is configured to search reference matrices of probabilities from the gesture library for matching with the matrix of probabilities, and to recognize the gesture eigenvalues as a reference gesture corresponding to the reference matrix of probabilities if the reference matrix of probabilities is found.

Claims

1. A gesture recognition apparatus for recognizing a gesture made by a person, comprising:

an ultrasound transmitter configured to send ultrasound signals;

a ultrasound receiver configured to receive reflected ultrasound signals from an object;

a dividing module configured to divide the reflected ultrasound signals into a plurality of frames according to a time interval;

a computing module configured to obtain an eigenvalue of each frame;

a classifying module configured to filter the eigenvalues to obtain gesture eigenvalues, and to obtain a matrix of probabilities of the gesture eigenvalues by using a sound recognition model;

a gesture library comprising a plurality of reference matrixes of probabilities corresponding to a plurality of reference gestures; and

a recognition module configured to search a reference matrix of probabilities from the gesture library for a match to the matrix of probabilities of the gesture eigenvalues; and

recognize the gesture eigenvalues as a reference gesture corresponding to the reference matrix of probabilities if the reference matrix of probabilities is found.

2. The gesture recognition apparatus of claim 1, further comprising an adjusting module configured to filter and amplify the reflected ultrasound signals.

3. The gesture recognition apparatus of claim 1, wherein the sound recognition model is a hidden markov model.

4. The gesture recognition apparatus of claim 1, wherein the recognition module configured to search the reference matrix of probabilities using a Viterbi search.

5. A gesture recognition method for recognizing a gesture made by a person, applied in an apparatus, comprising:

sending ultrasound signals using a ultrasound transmitter;

receiving reflected ultrasound signals using a ultrasound receiver;

dividing the reflected ultrasound signals into a plurality of frames according to a time interval;

obtaining an eigenvalue of each frame;

filtering the eigenvalues to obtain gesture eigenvalues and obtaining a matrix of probabilities of the gesture eigenvalues by using a sound recognition model; and

searching a reference matrix of probabilities for a match to the matrix of probabilities of the gesture eigenvalues, and recognizing the gesture eigenvalues as a reference gesture corresponding to the reference matrix of probabilities if the reference matrix of probabilities is found.

6. The gesture recognition method of claim 5, further comprising filtering and amplifying the reflected ultrasound signals before dividing the reflected ultrasound signals.

7. The gesture recognition method of claim 5, wherein the sound recognition model is a hidden markov model.

8. The gesture recognition method of claim 5, wherein searching the reference matrix of probabilities is using a Viterbi search.

9. A gesture recognition method, applied in an apparatus, comprising:

providing an ultrasound transmitter configured to send ultrasound signals;

an ultrasound receiver configured to receive reflected ultrasound signals from an object;

a dividing module configured to divide the reflected ultrasound signals into a plurality of frames according to a time interval;

a computing module configured to obtain an eigenvalue of each frame;

a classifying module configured to filter the eigenvalues to obtain gesture eigenvalues and to obtain a matrix of probabilities of the gesture eigenvalues by using a sound recognition model;

a gesture library comprising a plurality of reference matrixes of probabilities corresponding to a plurality of reference gestures; and

a recognition module configured to search a reference matrix of probabilities from the gesture library for a match to the matrix of probabilities of the gesture eigenvalues; and

recognize the gesture eigenvalues as a reference gesture corresponding to the reference matrix of probabilities if the reference matrix of probabilities is found;

sending ultrasound signals using the ultrasound transmitter, and receiving the reflected ultrasound signals using the ultrasound receiver;

dividing the reflected ultrasound signals into the plurality of frames according to the time interval;

obtaining the eigenvalue of each frame;

filtering the eigenvalues to obtain the gesture eigenvalues and obtaining the matrix of probabilities of the gesture eigenvalues by using the sound recognition model;

searching the reference matrix of probabilities for matching with the matrix of probabilities of the gesture eigenvalues, and recognizing the gesture eigenvalues as the reference gesture corresponding to the reference matrix of probabilities if the reference matrix of probabilities is found.