초록 ▼

A neural net sonic identification system includes an ultra-sonic emitter and receiver configured to detect characteristics of a subject by detecting reflections of the ultrasonic waves. The sonic reflections are processed by a neural network pattern recognition system that learns to recognize a part

A neural net sonic identification system includes an ultra-sonic emitter and receiver configured to detect characteristics of a subject by detecting reflections of the ultrasonic waves. The sonic reflections are processed by a neural network pattern recognition system that learns to recognize a particular subject and distinguish it from other subjects. The emitter/receiver is continuously optimized for phase, wavelength, spectral, and power frequencies, and the received sound “picture” is written-over the previously stored encrypted image data file for a particular individual, so as to continuously update the processor's retained identification information. In one embodiment, the emitter/receiver comprises a solid-state multi-layer tunable array of piezoelectric emitters and receivers that can be selectively “aimed” in different directions without physically moving by independently varying the wavelength and power of the various emitters/receivers.

대표청구항 ▼

1. A sonic identification system, comprising:a) an emitter, configured to broadcast ultra-sonic energy toward a subject at a variety of frequencies within a frequency range during a time interval; b) a receiver, configured to receive and detect the ultra-sonic energy reflected off of the subject, an

1. A sonic identification system, comprising:a) an emitter, configured to broadcast ultra-sonic energy toward a subject at a variety of frequencies within a frequency range during a time interval; b) a receiver, configured to receive and detect the ultra-sonic energy reflected off of the subject, and to develop signals representing characteristics of the reflected ultra-sonic energy; and c) a microprocessor, coupled to at least one of the emitter and receiver, including: i) a pre-processor, configured to receive the signals from the receiver and to develop a new spectrogram of the subject comprising a plurality of said signals taken over the time interval; ii) a memory, configured for containing records of spectrograms related to a plurality of known subjects; and iii) a processor configured to analyze the new spectrogram using a pattern-matching algorithm, to identify invariant features of the new spectrogram, and to determine whether the identity of the subject of the new spectrogram is the same as the subject of any of the records of spectrograms stored in the memory. 2. A system in accordance with claim 1, wherein the frequency range includes frequencies that penetrate soft tissue, and frequencies that reflect off of soft tissue of a human subject.3. A system in accordance with claim 1, wherein the frequency range is from about 45 kHz to about 100 kHz.4. A system in accordance with claim 1, wherein the emitter is configured to continuously vary the frequency across the frequency range during the time interval.5. A system in accordance with claim 1, wherein the emitter is configured to emit ultra-sonic energy at a variety of power levels during the time interval.6. A system in accordance with claim 5, wherein the variety of power levels are within a range of from about 0.25 watts to about 2.5 watts.7. A system in accordance with claim 1, wherein the emitter is a piezo-electric transducer.8. A system in accordance with claim 1, wherein the emitter and receiver comprise an array of emitters and receivers disposed upon a common support structure.9. A system in accordance with claim 8, wherein the common support structure comprises a back plate and a front plate, and the directional orientation of the array can be changed by varying the wavelength and frequency of signals transmitted to emitters on the back plate and front plate relative to each other, without physically moving the array.10. A system in accordance with claim 9, wherein the microprocessor is configured to process the received signals using a Fourier transform algorithm, and to tune the sensor array based upon a recognition refinement algorithm by selectively controlling (a) an on/off condition of each of the emitters and receivers in the array; (b) the frequency and wavelength of ultra-sonic energy broadcast by the emitters; and (c) the directional orientation of the array, so as to increase clarity and reduce noise in the signals developed by the receivers.11. A system in accordance with claim 1, wherein the microprocessor is configured to compare the new spectrogram to the spectrographic information related to the other subjects using a neural net pattern recognition algorithm, so as to determine a likelihood of correlation of the new spectrogram to the other spectrograms, and thereby allow identification of the subject as being identical to the subject of another spectrogram if the likelihood of correlation is above a selected threshold level.12. A system in accordance with claim 11, wherein the microprocessor is configured to:a) update its memory by overwriting the previously-stored spectrogram related to the subject with the new spectrogram upon identification of the new spectrogram as being identical to another spectrogram, so as to increase the likelihood of future identification of the subject; and b) store the new spectrogram separately in memory if unable to identify the new spectrogram as being identical to another spectrogram above the threshold level. 13. A system in accordance with claim 1, wherein the microprocessor is configured to continuously develop a plurality of spectrograms over the time interval, and to store the plurality of spectrograms in memory, so as to retain spectrographic information related to the subject in a variety of conditions.14. A system in accordance with claim 13, wherein the variety of conditions include varying spatial orientations of the subject, and varying physical features.15. A system in accordance with claim 1, wherein the receiver is configured to develop time-domain signals representing the characteristics of the reflected ultra-sonic energy, and the pre-processor is configured to convert the time-domain signals to frequency domain signals for creating the new spectrogram.16. A sonic identification system, comprising:a) an array of ultra-sonic emitters and receivers, the emitters being configured to broadcast ultra-sonic energy toward a subject at a variety of frequencies within a frequency range over a time interval, and the receivers being configured to receive and detect ultra-sonic energy reflected off of the subject, and to develop signals representing characteristics of the reflected ultra-sonic energy; and b) a microprocessor, having a memory, configured to: i) receive the signals from the receiver; ii) develop a new spectrogram of signals reflected from the subject over the time interval; iii) compare the new spectrogram of the subject with other spectrographic information stored in memory, using a neural net pattern recognition algorithm, so as to determine a likelihood of correlation of the new spectrogram with the other spectrographic information; iv) identify the subject as being identical to the subject of another spectrogram if the likelihood of correlation is above a selected threshold level; and v) update the spectrographic information stored in memory by replacing the previously-stored spectrographic information related to the subject with the new spectrogram, thereby increasing the likelihood of future identification of the subject. 17. A method of identifying characteristics of a subject, comprising the steps of:a. exposing a first subject to ultra-sonic energy at a variety of frequencies within a frequency range over a time interval; b. receiving and detecting reflections of the ultra-sonic energy from the first subject; c. producing a spectrogram representing the reflections over the time interval using a microprocessor having a memory containing a plurality of prior records of spectrographic information, each of the plurality of prior records being related to a specific subject; d. comparing the spectrogram of the first subject to each of the records of spectrographic information contained in the memory using a pattern recognition algorithm, so as to determine a likelihood of correlation of the first spectrogram to any of the plurality of prior records; e. identifying the first subject as being identical to the subject related to a prior record if the likelihood of correlation is above a selected threshold level; and f. storing the spectrogram related to the first subject in memory for future comparison with other spectrograms. 18. A method in accordance with claim 17, further comprising the steps of:g) repeating steps a-d for a second subject; h) comparing the spectrogram of the second subject to the spectrogram of the first subject using a pattern recognition algorithm, so as to determine a likelihood of correlation of the first spectrogram to the second spectrogram; and i) identifying the second subject as being the first subject if the likelihood of correlation is above a selected threshold level. 19. A method in accordance with claim 17, wherein the step of producing the spectrogram comprises the more specific step of analyzing the reflections using a Fourier transform algorithm.20. A method in accordance with claim 17, wherein the step of storing the spectrogram in the memory comprises replacing the previously-stored prior record related to the first subject with a new record related to the first subject, the new record comprising the spectrogram, such that the microprocessor retains in memory the most current spectrographic information related to the first subject.21. A method in accordance with claim 17, wherein the step of storing the spectrogram in the memory comprises creating a new record of spectrographic information that combines at least a portion of the previously-stored prior record related to the first subject with at least a portion of the spectrogram, such that the microprocessor retains in memory a broad profile of spectrographic information related to the first subject.22. A method in accordance with claim 17, wherein subject is a person.23. A method in accordance with claim 22, wherein the characteristics of the subject are selected from the group consisting of the person's facial appearance, the orientation of the person's physical body, and the person's actions.24. A method for a user to actuate a device, comprising the steps of:a. exposing the user to ultra-sonic energy at a variety of frequencies within a frequency range over a time interval; b. receiving and detecting reflections of the ultra-sonic energy from the user; c. producing a new spectrogram representing the reflections over the time interval using a microprocessor having a memory containing a plurality of prior records of spectrographic information, each of the plurality of prior records being related to an authorized user of the device; d. comparing the spectrogram of the user to each of the prior records of spectrographic information contained in the memory using a pattern recognition algorithm, so as to determine a likelihood of correlation of the new spectrogram to any of the plurality of prior records; e. identifying the first subject as being an authorized user of the device if the likelihood of correlation is above a selected threshold level; f. storing the new spectrogram in memory for future comparison with other spectrograms; and g. actuating the device. 25. A method in accordance with claim 24, wherein the device is selected from the group consisting of an access lock, and a transport vehicle.26. A method in accordance with claim 25, wherein the access lock is selected from the group consisting of a door lock, a padlock, and a safe lock.27. A method in accordance with claim 25, wherein the transport vehicle is selected from the group consisting of an automobile, a truck, a motorcycle, an airplane, and a boat.28. A method in accordance with claim 24, wherein the step of storing the spectrogram in the memory comprises replacing the previously-stored prior record related to the user with a new record related to the user, the new record comprising the new spectrogram, such that the microprocessor retains in memory the most current spectrographic information related to the user.29. A method in accordance with claim 24, wherein the step of storing the spectrogram in the memory comprises creating a new record of spectrographic information that combines at least a portion of the previously-stored prior record related to the user with at least a portion of the new spectrogram, such that the microprocessor retains in memory a broad profile of spectrographic information related to the user.