Drone detection and classification methods and apparatus
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
G10L-025/51
G10L-025/54
G10L-025/18
G01S-003/80
G01H-001/00
H04R-029/00
G01S-005/18
G10L-019/00
출원번호
US-0950593
(2015-11-24)
등록번호
US-9697850
(2017-07-04)
발명자
/ 주소
Hearing, Brian
Franklin, John
출원인 / 주소
Droneshield, LLC
대리인 / 주소
K&L Gates LLP
인용정보
피인용 횟수 :
0인용 특허 :
12
초록▼
A system, method, and apparatus for drone detection and classification are disclosed. An example method includes receiving a sound signal in a microphone and recording, via a sound card, a digital sound sample of the sound signal, the digital sound sample having a predetermined duration. The method
A system, method, and apparatus for drone detection and classification are disclosed. An example method includes receiving a sound signal in a microphone and recording, via a sound card, a digital sound sample of the sound signal, the digital sound sample having a predetermined duration. The method also includes processing, via a processor, the digital sound sample into a feature frequency spectrum. The method further includes applying, via the processor, broad spectrum matching to compare the feature frequency spectrum to at least one drone sound signature stored in a database, the at least one drone sound signature corresponding to a flight characteristic of a drone model. The method moreover includes, conditioned on matching the feature frequency spectrum to one of the drone sound signatures, transmitting, via the processor, an alert.
대표청구항▼
1. An apparatus for creating a drone signature library comprising: a microphone configured to receive a sound signal from a drone;a sound card configured to record a digital sound sample of the sound signal; anda sample processor configured to: partition the digital sound sample into a predetermined
1. An apparatus for creating a drone signature library comprising: a microphone configured to receive a sound signal from a drone;a sound card configured to record a digital sound sample of the sound signal; anda sample processor configured to: partition the digital sound sample into a predetermined number of segments having a specified duration,convert each of the segments into a vector of frequency amplitudes by applying a frequency domain transformation to the segments,form a composite frequency vector by averaging the vectors,determine a drone component within the composite frequency vector,receive drone information indicative of a type of the drone,associate the drone information with the drone component of the composite frequency vector,store information related to the drone component of the composite frequency vector and the drone information to the drone signature library, anduse the information related to the drone component and the drone information for detection of drones. 2. The apparatus of claim 1, wherein the sound signal is generated by at least one of a sound machine, a user device operating an application configured to cause the user device to generate the sound signal, and a drone. 3. The apparatus of claim 1, wherein the identifier specifies at least one of a drone type, a drone class, and a drone flight characteristic. 4. The apparatus of claim 3, wherein the drone flight characteristic includes at least one of retreating, advancing, sideways translating, rotating, hovering, inverting, ascending, and descending; and wherein the drone type includes at least one of a brand name, a model name, a part number, and a rotor configuration. 5. The apparatus of claim 1, wherein the calibration processor is configured to: determine at least one of (i) tuning parameters, and (ii) calibration filters for different sound signals;combine the at least one of (i) and (ii) for the different sound signals into at least one of an average tuning parameter and an average calibration filter; andapply the at least one of an average tuning parameter and an average calibration filter to the sample processor. 6. The apparatus of claim 5, wherein the different sound signals are generated by a sound machine according to a predetermined routine and are configured to account for particular tones. 7. The apparatus of claim 1, wherein the at least one of the tuning parameter and the calibration filter is configured to at least one of: shift a frequency of a subsequent digital sound sample,filter a component from a subsequent digital sound sample,shift a phase of a subsequent digital sound sample,shift a frequency peak within a subsequent digital sound sample or a subsequent measured composite frequency vector, andsmooth a frequency peak within a subsequent digital sound sample or a subsequent measured composite frequency vector. 8. The apparatus of claim 1, wherein the calibration processor is configured to be activated after the microphone is placed in a location to detect drones. 9. The apparatus of claim 1, wherein the calibration processor is configured to receive from a management server, via a network, the calibration frequency vector. 10. The apparatus of claim 1, wherein the sample processor is configured to: smooth the vectors of each of the segments using a filter; andform the measured composite frequency vector by averaging the smoothed vectors. 11. The apparatus of claim 1, wherein the predetermined number of segments are equal-sized and non-overlapping. 12. A drone detection apparatus comprising: a microphone configured to receive a sound signal that is related to a drone;a sound card configured to record a measured digital sound sample of the sound signal; anda calibration processor configured to: receive an identifier that is related to the digital sound sample,select a calibration digital sound sample stored in a memory that corresponds to the received identifier,determine differences between the measured digital sound sample and the calibration digital sound sample,determine a tuning parameter based on the determined differences, andcalibrate a sample processor using the tuning parameter for identifying a drone from subsequent sound signals recorded by the microphone. 13. The apparatus of claim 12, wherein at least one of: (i) the sound signal includes background noise, and(ii) the sound signal is affected by an environmental characteristic. 14. The apparatus of claim 13, wherein the calibration processor is configured to calibrate the sample processor to at least one of compensate for the background noise and compensate for the environmental characteristic. 15. The apparatus of claim 12, wherein the calibration processor is configured to calibrate the sample processor by causing the sample processor to apply the tuning parameter to the subsequent sound signals. 16. The apparatus of claim 12, wherein the calibration processor is configured to calibrate the sample processor by causing the sample processor to apply the tuning parameter to frequency signatures related to drones. 17. The apparatus of claim 12, wherein the calibration processor is configured to: determine a calibration filter based on the determined differences between the measured digital sound sample and the calibration digital sound sample; andcalibrate the sample processor using the calibration filter. 18. A method for calibrating a drone detection device comprising: (i) receiving, in a microphone, a sound signal that is representative of a drone;(ii) recording, via a sound card, a digital sound sample of the sound signal;(iii) partitioning, via a processor, the digital sound sample into a predetermined number of segments having a specified duration;(iv) converting, via the processor, each of the segments into a vector of frequency amplitudes by determining an absolute value of a Fast Fourier Transform applied to the respective segment;(vi) receiving, in the processor, an identifier that is related to the sound signal;(vii) selecting, via the processor, a calibration frequency vector stored in a memory that corresponds to the received identifier;(viii) determining, via the processor, a difference between the calibration frequency vector and each of the vectors of frequency amplitudes;(ix) determining, via the processor, a tuning parameter based on the determined differences; and(x) applying, via the processor, the tuning parameter for compensation when sound signals are recorded by the microphone. 19. The method of claim 18, wherein step (iv) includes: smoothing, via the processor, the vectors of each of the segments using a filter, andaveraging, via the processor, the smoothed vectors to form a measured composite frequency vector;wherein step (viii) includes determining, via the processor, a difference between the calibration frequency vector and the measured composite frequency vector; andwherein step (ix) includes determining, via the processor, the tuning parameter based on the determined difference between the calibration frequency vector and the measured composite frequency vector. 20. The method of claim 18, wherein steps (i) to (ix) are repeated for at least one of different flight characteristics of a drone and different types of drones, and wherein the step (x) further includes: averaging the tuning parameter based on each iteration of steps (i) to (ix), andapplying the averaged tuning parameter for the detection of drones. 21. The method of claim 20, wherein the sound signal for each iteration of steps (i) to (ix) is generated according to a predetermined routine and includes predetermined tones known to be affected by environmental characteristics. 22. The method of claim 21, wherein the predetermined routine is operated by an application operating on a user device, and wherein the method further comprises, for each sound signal within the predetermined routine, transmitting the corresponding identifier from the application to the processor. 23. The method of claim 18, wherein the identifier is generated acoustically and provided in conjunction with the sound signal. 24. The method of claim 18, wherein the sound signal is representative of a sound emitted by a drone.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (12)
Lee, Jonathan S.; Nielsen, Richard O.; Davis, Henry Hooper; Dunholter, Paul H., Acoustic airport surveillance system.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.