Method and apparatus for the automatic identification of birds by their vocalizations

• US 20060150920A1
• Filed: 01/11/2005
• Published: 07/13/2006
• Est. Priority Date: 01/11/2005
• Status: Active Grant

• Current Assignee: Pariff LLC
• Original Assignee: Pariff LLC
• Inventors: Patton, Charles M.
• Granted Patent: US 7,377,233 B2
• Time in Patent Office: Days
• US Class Current: 119/712
• CPC Class Codes: A01K 11/008 incorporating GPS
• CPC Class Codes: A01K 35/00 Marking poultry or other birds marking meat A22C17/10
• CPC Class Codes: A01K 45/00 Other aviculture appliances, e.g. devices for determining whether a bird is about to lay
• CPC Class Codes: G09B 19/06 Foreign languages with audible presentation of material to be studied G09B5/04
• CPC Class Codes: G09B 23/36 for zoology

Abstract

An apparatus for detecting and identifying birds based upon electronic analysis of their bird calls and songs and method for doing so by utilizing a step-by-step hierarchical method of breaking down bird vocalizations according to order, family, and species of the specific bird. Several embodiments of the apparatus are disclosed particularly a hand held computational device, microphone, audio capture card, user application software and a collection of prerecorded audio data.

Claims

1. A method for identifying birds comprising the steps of:

detecting a sound segment;

determining if the sound segment detected is the sound of a bird singing;

determining the family of the bird by analyzing the dynamical modes of the sound segment and thereafter;

determining the species of the bird by comparing audio parameters characteristic of species within the family to the detected bird sound.

2. The method of claim 1 wherein the step of detecting if the sound is the sound of a bird includes the step of detecting the presence of dual vibrating membranes.

3. The method of claim 2 wherein the step of analyzing the dynamical modes of the sound segment includes the step of determining the audio characteristics of the dynamical modes.

4. The method of claim 3 wherein the step of determining the audio characteristics include the matching of the detected audio characteristics with audio characteristics of a specific family of birds.

5. The method of claim 4 wherein the step of comparing audio parameters with detected bird sounds includes the step of detecting a patterned sequence of shifts between dynamical modes to identify the specific species of the bird.

6. A method for detecting the presence of a bird in a pre-determined area comprising the steps of:

detecting an audio signal characteristic of a bird dual membrane sound at a first detector;

communicating the detected bird audio signal to a second detector for identifying the dynamical mode characteristic of a particular family of bird; and

thereafter communicating the detected bird audio signal to a third detector for determining the species of bird within the family of bird previously identified.

7. The method of claim 6 wherein the step of detecting employs a sound pickup device.

8. The method of claim 7 wherein the step of communicating the detected signal includes the use of an audio capture subsystem.

9. The method of claim 8 wherein the step of identifying the dynamical mode includes the step of comparing the detected audio signal to audio signals characteristics of a particular family of birds.

10. The method of claim 9 wherein the step of determining the species of a bird includes the step of comparing the detected audio signal to audio signals characteristic of a particular species within the family.

11. An apparatus for identifying the species of a bird comprising:

sound detecting device;

an audio capture sub-system operatively connected to the sound detection device;

a computational device operatively connected to the audio capture sub-system, the computational device including a software program for operating the computational device, and;

a collection of audio data pre-recorded in the audio capture card for determining the family and species of a bird detected by the apparatus.

12. An apparatus according to claim 11 wherein the sound detecting device is a microphone.

13. An apparatus according to claim 12 wherein the computational device is a personal digital assistant.

14. An apparatus according to claim 13 wherein the audio capture card is operatively connected to the personal digital assistant.

15. An apparatus according to claim 14 wherein the computational device compares detailed audio data to the audio data collection on the audio card.

16. An apparatus according to claim 11 wherein the microphone is attached to a diaphragm to enhance the sensitivity of the system.

17. An apparatus according to claim 16 wherein the diaphragm is a window.

18. An apparatus according to claim 11 wherein the microphone is attached at a point remote from the computational device.

19. An apparatus according to claim 18 wherein the computational device is a conventional computer.

20. An apparatus for classifying portions of audio signals, especially signals produced by natural sources, comprising:

a first signal property receiver for detecting selectable portions of the signal and selecting one of a predetermined set of outcomes;

a signal scanner operatively connected to the receiver for recording the results at locations that are correlated to the portions of the signal to which the detection means were applied;

a second signal property detector to apply to each region of the signal based on the outcomes of the first signal property detection means recorded at positions correlated to portions of the signal in and adjacent to the region;

means for scanning said signal, successively applying said chosen detection means, and recording the results at locations which are correlated to the portions of the signal to which the detection means were applied;

wherein based on the recorded outcomes of the second signal property detector, one or more subsets of the signal are chosen and compared in a third signal property detector with outcomes in a desired set of classifications; and

thereafter recording the results in locations that are correlated with the portions of the signal that comprise the subsets.

21. A system for classifying portions of recorded signals, especially signals produced by natural sources, comprising:

a first signal property detection means applicable to selectable portions of the signal whose result upon each application is one of a predetermined set of outcomes;

means for scanning said signal, successively applying said detection means, and recording the results at locations that are correlated to the portions of the signal to which the detection means were applied;

a second signal property detection means by comparing each region of the signal based on the outcomes of the first signal property detection means recorded at positions correlated to portions of the signal in and adjacent to the region;

means for scanning said signal, successively applying said chosen detection means, and recording the results at locations which are correlated to the portions of the signal to which the detection means were applied;

wherein, based on the recorded outcomes of the second signal property detection means, one or more subsets of the signal are chosen and compared in a third signal property detection means with outcomes in a desired set of classifications; and

thereafter recording the results in locations that are correlated with the portions of the signal that comprise the subsets.

22. A method of identifying animal sounds, comprising:

recording a sound produced by an animal as audio data,

analyzing said audio data to determine a first hierarchy of sounds,

analyzing the first hierarchy to determine the family of the animal,

comparing the first hierarchy to predetermined audio data of various species of animal to determine the species of the animal.

23. The method of claim 22, wherein the analyzing step comprises analyzing the dynamical modes present in the audio data and comparing the dynamical modes with a stored set of dynamical modes characteristic of the family of the animal.

24. A method for learning to identify birds by their song comprising:

storing a bird'"'"'s song in a memory device;

storing audio feature relevance data temporarily correlated to said bird'"'"'s song in a memory device;

selecting the stored bird'"'"'s song for replay; and

replaying the selected bird'"'"'s song on an audio device while displaying the correlated relevance data on a display device.

25. The method of claim 24, wherein the step of storing audio relevance data includes storing audio family index data and audio species identification data.

26. The method of claim 24, wherein the step of replaying the selected song and displaying the correlated data on a display device includes displaying the degree of relevance of the data to the identification of the bird family.

27. The method of claim 24, wherein the step of replaying the selected song and displaying the correlated data on a display device includes displaying the degree of relevance of the data to the identification of the bird species.

28. The method of claim 27, including the step of controlling the replay of the stored bird'"'"'s song to select portions of the song with high relevance to the bird'"'"'s species.

29. The method of claim 28, including the step of selecting a specific time point of the stored song.

30. The method of claim 29, including the step of displaying the relevance of the selected time point to the identification of the family and species of the bird.

31. The method of claim 30 including the step of displaying the relevance of the selected time point to the identification of the species of the bird.