Automatic gesture recognition for a sensor system
A method for gesture recognition including detecting one or more gesture-related signals using the associated plurality of detection sensors; and evaluating a gesture detected from the one or more gesture-related signals using an automatic recognition technique to determine if the gesture corresponds to one of a predetermined set of gestures.
1. A method for touchless gesture recognition comprising:
detecting one or more gesture-related signals using an associated plurality of detection sensors; and
evaluating the touchless gesture detected from the one or more gesture-related signals using an automatic recognition technique to determine if the touchless gesture corresponds to one of a predetermined set of gestures, wherein in determining a start of the gesture, a start is determined if the distance between the target object and at least one sensor decreases and the distance between the target object and at least another sensor increases, and a short term variance or an equivalent measure over a predetermined plurality of signal samples is less than a threshold.
2. A method in accordance with claim 1 , wherein evaluating the gesture includes determining a stop of a gesture.
3. A method in accordance with claim 1 , wherein a stop of a the gesture is determined if at a given time the distances between the target object and all sensors decrease, and/or a short term variance or an equivalent measure over a predetermined plurality of signal samples is less than a threshold, and/or the signal changes are less than a predetermined threshold for a predetermined plurality of signal samples after the given time.
4. A method in accordance with claim 1 , wherein each gesture is represented by one or more Hidden Markov Models (HMM).
5. A method in accordance with claim 4 , wherein evaluating a gesture includes evaluating probability measures for one or more HMMs.
6. A method in accordance with claim 4 , wherein features to which observation matrices of the one or more HMMs are associated are non-quantized or quantized sensor signal levels, x/y/z position, distances, direction, orientation, angles and/or 1 st , 2 nd or higher order derivatives of these with respect to time, or any combination thereof.
7. A method in accordance with claim 4 , wherein features to which observation matrices of the one or more HMMs are associated are the 1 st derivatives of sensor signal levels quantized to two quantization levels.
8. A method in accordance with claim 5 , wherein for each new signal sample or feature, the probability of each Hidden Markov Model is updated.
9. A method in accordance with claim 8 , wherein if the probability of a Hidden Markov Model exceeds a pre-defined threshold, the recognition is stopped.
10. A system for gesture recognition using an alternating electric field generated by a sensor arrangement and associated detection electrodes, wherein a gesture is performed without touching a surface, wherein electrode signals are evaluated using Hidden Markov Models, wherein start and stop criteria for determination of a gesture are determined, and wherein feature sequences used to evaluate the Hidden Markov Models' probabilities are the 1 st derivatives of sensor signal levels quantized to two quantization levels.
11. A system for gesture recognition comprising: a sensor arrangement for detecting one or more gesture-related signals using an associated plurality of detection sensors; and a module for evaluating a touchless gesture detected from the one or more gesture-related signals using an automatic recognition technique to determine if the gesture corresponds to one of a predetermined set of gestures, wherein a start of the gesture is determined if the distance between the target object and at least one sensor decreases and the distance between the target object and at least one sensor increases and a short term variance or an equivalent measure over a predetermined plurality of signal samples is less than a threshold.
12. A system in accordance with claim 11 , wherein a stop of the gesture is determined if at a given time, the distances between the target object and all sensors decrease and/or a short term variance or an equivalent measure over a predetermined plurality of signal samples is less than a threshold, and/or the signal changes are less than a predetermined threshold for a predetermined plurality of signal samples after the given time.
13. A system in accordance with claim 11 , wherein each gesture is represented by one or more Hidden Markov Models.
14. A system in accordance with claim 13 , wherein evaluating a gesture includes evaluating probability measures for one or more Hidden Markov Models.
15. A system in accordance with claim 13 , wherein the features to which observation matrices of the one or more HMMs are associated are non-quantized or quantized sensor signal levels, x/y/z position, distances, direction, orientation, angles and/or 1 st , 2 nd or higher order derivatives of these with respect to time, or any combination thereof.
16. A system in accordance with claim 13 , wherein the features are the 1 st derivatives of the sensor signal levels quantized to two quantization levels.
17. A system in accordance with claim 14 , wherein for each new signal sample or feature, the probability of each Hidden Markov Model is updated.
18. A system in accordance with claim 17 , wherein if the probability of a Hidden Markov Model exceeds a pre-defined threshold, the recognition is stopped.
19. A computer readable medium including one or more non-transitory machine readable program instructions for receiving one or more gesture-related signals using a plurality of detection sensors; and evaluating a touchless gesture detected from the one or more gesture-related signals using an automatic recognition technique to determine if the gesture corresponds to one of a predetermined set of gestures, wherein a start of a gesture is determined if the distance between the target object and at least one sensor decreases and the distance between the target object and at least one other sensor increases, and a short term variance or an equivalent measure over a predetermined plurality of signal samples is less than a threshold.
20. A computer readable medium in accordance with claim 19 , wherein a stop of the gesture is determined if at a given time, the distances between the target object and all sensors decrease and/or a short term variance or an equivalent measure over a predetermined plurality of signal samples is less than a threshold, and/or the signal changes are less than a predetermined threshold for a predetermined plurality of signal samples after the given time.
21. A computer readable medium in accordance with claim 19 , wherein each gesture is represented by one or more Hidden Markov Models (HMM).
22. A computer readable medium in accordance with claim 21 , wherein evaluating a gesture includes evaluating a probability measure for one or more HMMs.
23. A computer readable medium in accordance with claim 21 , wherein features to which observation matrices of the one or more HMMs are associated are non-quantized or quantized sensor signal levels, x/y/z position, distances, direction, orientation, angles and/or 1 st , 2 nd or higher order derivatives of these with respect to time, or any combination thereof.
24. A computer readable medium in accordance with claim 21 , wherein the features are the 1 st derivatives of the sensor signal levels quantized to two quantization levels.
25. A computer readable medium in accordance with claim 22 , wherein for each new signal sample or feature, the probability of each Hidden Markov Model is updated.
26. A computer readable medium in accordance with claim 25 , wherein if the probability of Hidden Markov Model exceeds a pre-defined threshold, the recognition is stopped.
27. A system for gesture recognition comprising:
a sensor arrangement for detecting one or more gesture-related signals using an associated plurality of detection sensors; and
a module for evaluating a touchless gesture detected from the one or more gesture-related signals using an automatic recognition technique to determine if the gesture corresponds to one of a predetermined set of gestures, wherein each gesture is represented by one or more Hidden Markov Models, and wherein features to which observation matrices of the one or more HMMs are associated are the 1 st derivatives of sensor signal levels quantized to two quantization levels.