Method and apparatus for processing signals to extract narrow bandwidth features

• US 5,245,589 A
• Filed: 03/20/1992
• Issued: 09/14/1993
• Est. Priority Date: 03/20/1992
• Status: Expired due to Fees

• Current Assignee: Ho J. Lee, Jonathan S. Abel
• Original Assignee: Ho J. Lee, Jonathan S. Abel
• Inventors: Lee, Ho J., Abel, Jonathan S.
• Primary Examiner(s): Pihulic, Daniel T.
• Application Number:
  US07/856,091
• Time in Patent Office: 543 Days
• Field of Search: 367/87, 367/89, 367/100, 367/124, 367/125, 367/901, 367/135, 367/136, 364/574, 381/94, 381/71, 342/192, 342/195, 342/203
• US Class Current: 367/136
• CPC Class Codes: G01R 23/16 Spectrum analysis; Fourier analysis

Abstract

A method and apparatus for processing a set of signals to identify narrow bandwidth features of the signals, and optionally to process the signals further to extract information about the identified narrow-band features. The invention processes a set of input signal frames (a two-dimensional pixel array) to generate a narrow-band feature signal (also a two-dimensional pixel array) from which narrow-band features of the input signal frames can be efficiently, automatically, and unambiguously identified. In a class of preferred embodiments, the input signal frames are the power spectra of a set of sequentially measured signals. Thus, the set of input signal frames is a "spectrogram," comprising rows and columns of pixels (with row indices representing time, and column indices representing frequency). Alternatively, the input signal frames represent a data array of another type, such as a correlogram or a sequence of images. In a class of embodiments, the input signal frames are processed to identify narrow-band pixels (the subset of input signal frame pixels which represent narrow-band energy, or narrow-band processes). Then, the identified narrow-band pixels (which can be displayed as a narrow-band pixel array) undergo "feature extraction" processing to generate the output narrow-band feature signal (which can be displayed as the output image). The narrow-band feature signal can be further processed to determine the center frequency, bandwidth, and amplitude of one or more of its narrow-band features.

Claims

1. A method for processing a set of input signals, including the steps of:

generating an array of input pixels from the input signals, wherein each of the input pixels has a time coordinate and a column coordinate; and

processing the input pixels to generate a classification pixel array of classification pixels corresponding to the input pixels, wherein the classification pixels identify a first subset of the input pixels representing narrow-band energy and a second subset of the input pixels representing broad-band energy.

2. The method of claim 1, also including the step of:

generating from the array of input pixels a narrow-band feature signal comprising narrow-band pixels, and uniform pixels having substantially uniform value, by replacing the input pixels belonging to the second subset with said uniform pixels, wherein the narrow-band pixels of the narrow-band feature signal define a set of narrow-band features.

3. The method of claim 2, also including the step of:

displaying the narrow-band feature signal as an image comprising the narrow-band features on a contrasting substantially uniform background.

4. The method of claim 3, also including the step of:

generating a feature parameter signal indicative of at least one parameter of at least one narrow-band feature in the set of narrow-band features.

5. The method of claim 1, wherein the column coordinate is a frequency coordinate.

6. The method of claim 5, wherein the array of input pixels is a spectrogram comprising rows of pixels, wherein each of the rows defines a power spectrum of one of the input signals.

7. The method of claim 1, wherein the column coordinate is a time interval coordinate.

8. The method of claim 7, wherein the array of input pixels is a correlogram comprising rows of pixels, wherein each of the rows defines an auto-correlation of one of the input signals.

9. The method of claim 7, wherein the array of input pixels is a correlogram comprising rows of pixels, wherein each of the rows defines a cross-correlation of two of the input signals.

10. The method of claim 1, wherein the column coordinate is a spatial coordinate.

11. The method of claim 10, wherein each of the input signals is a set of first pixels defining an image, and wherein the array of input pixels consists of a subset of the first pixels.

12. The method of claim 1, wherein the input signals are analog signals, and wherein the step of generating the array of input pixels from the input signals includes the steps of:

generating digital data by digitizing the analog signals; and

processing the digital data to generate the array of input pixels.

13. The method of claim 1, wherein the step of processing the input pixels to generate the classification pixel array includes the steps of:

performing a broad-band equalization operation on the input pixels to generate equalized pixels; and

generating a narrow-band pixel array by performing a generalized likelihood ratio operation on the equalized pixels.

14. The method of claim 1, also includes the steps of:

performing feature extraction operations on the classification pixel array.

15. The method of claim 14, wherein the feature extraction operations include a dilation operation.

16. The method of claim 14, wherein the feature extraction operations include an erosion operation.

17. A method for processing a set of input signals, including the steps of:

generating the input signals at an acoustic receiver, in response to reception of acoustic waves;

digitizing each of the input signals to generate digital data, and generating power spectra of the digital data;

generating an array of input pixels from the power spectra, wherein each of the input pixels has a time coordinate and a frequency coordinate; and

processing the input pixels to generate a classification pixel array of classification pixels corresponding to the input pixels, wherein the classification pixels identify a first subset of the input pixels representing narrow-band energy and a second subset of the input pixels representing broad-band energy.

18. The method of claim 17, wherein the acoustic receiver is a hydrophone and the acoustic waves are sonar signals.

19. The method of claim 17, wherein the acoustic receiver is a microphone mounted at an airport and the acoustic waves are generated by aircraft in operation near the microphone.

34. The method of claim 17, wherein the acoustic receiver is a microphone and the acoustic waves are generated by a musical instrument.

20. A method for processing a set of input signals, including the steps of:

generating the input signals at at least one hydrophone, in response to reception of sonar signals;

digitizing each of the input signals to generate digital data, and performing a correlation operation on the digital data to generate correlation data for each of the input signals;

generating an array of input pixels from the correlation data, wherein each of the input pixels has a time coordinate and a time interval coordinate; and

processing the input pixels to generate a classification pixel array of classification pixels corresponding to the input pixels, wherein the classification pixels identify a first subset of the input pixels representing narrow-band energy and a second subset of the input pixels representing broad-band energy.

21. The method of claim 20, wherein the correlation operation is an auto-correlation operation.

22. The method of claim 20, wherein the correlation operation is a cross-correlation operation.

23. An apparatus for processing a set of input signals, including:

means for generating an array of input pixels from the input signals, wherein each of the input pixels has a time coordinate and a column coordinate; and

means for processing the input pixels to generate a classification pixel array of classification pixels corresponding to the input pixels, wherein the classification pixels identify a first subset of the input pixels representing narrow-band energy and a second subset of the input pixels representing broad-band energy.

24. The apparatus of claim 23, wherein the means for processing the input pixels also includes:

means for generating from the array of input pixels a narrow-band feature signal comprising narrow-band pixels, and uniform pixels having substantially uniform value, by replacing the input pixels belonging to the second subset with said uniform pixels, wherein the narrow-band pixels of the narrow-band feature signal define a set of narrow-band features.

25. The apparatus of claim 23, also including:

means for displaying the narrow-band feature signal as an image comprising the narrow-band features on a contrasting substantially uniform background.

26. The apparatus of claim 23, wherein the means for processing the input pixels also includes:

means for generating a feature parameter signal indicative of at least one parameter of at least one narrow-band feature in the set of narrow-band features.

27. The apparatus of claim 23, wherein the means for processing the input pixels also includes:

means for generating a first feature parameter signal indicative of at least one parameter of at least one narrow-band feature in the set of narrow-band features; and

look-up table means for asserting a second feature parameter signal in response to the first feature parameter signal.

28. The apparatus of claim 23, wherein the means for processing the input pixels includes:

means for performing a broad-band equalization operation on the input pixels to generate equalized pixels, and for generating a narrow-band pixel array by performing a generalized likelihood ratio operation on the equalized pixels.

29. The apparatus of claim 23, wherein the means for processing the input pixels includes:

means for performing feature extraction operations on the classification pixel array.

30. The apparatus of claim 29, wherein the feature extraction operations include a dilation operation and an erosion operation, wherein the means for performing feature extraction operations on the classification pixel array includes:

an averaging filter for receiving the classification pixels and generating averaged classification pixels by averaging selected groups of the classification pixels; and

threshold comparison means for comparing the averaged classification pixels to a dilation threshold signal and performing the dilation operation on the averaged classification pixels as a result of said comparison, and for comparing the averaged classification pixels to an erosion threshold signal and performing the erosion operation on the averaged classification pixels as a result of said comparison.

31. The apparatus of claim 23, wherein the column coordinate is a spatial coordinate.

32. An apparatus for processing a set of input signals, including:

an acoustic receiver for generating the input signals in response to reception of acoustic waves;

means for digitizing each of the input signals to generate digital data;

means for generating power spectra of the digital data;

means for generating an array of input pixels from the power spectra, wherein each of the input pixels has a time coordinate and a frequency coordinate; and

means for processing the input pixels to generate a classification pixel array of classification pixels corresponding to the input pixels, wherein the classification pixels identify a first subset of the input pixels representing narrow-band energy and a second subset of the input pixels representing broad-band energy.

33. An apparatus for processing a set of input signals, including:

a set of one or more acoustic receivers for generating the input signals response to reception of acoustic waves;

means for digitizing each of the input signals to generate digital data;

means for performing a correlation operation on the digital data to generate correlation data for each of the input signals;

means for generating an array of input pixels from the correlation data, wherein each of the input pixels has a time coordinate and a time interval coordinate; and

means for processing the input pixels to generate a classification pixel array of classification pixels corresponding to the input pixels, wherein the classification pixels identify a first subset of the input pixels representing narrow-band energy and a second subset of the input pixels representing broad-band energy.