Acoustic signature recognition and identification

Details Acoustic signature recognition and identification Acoustic signature recognition and identification

1997-09-30

2001-01-09

Mehta, Bhavesh (Department: 2721)

Image analysis

Pattern recognition

Feature extraction

C382S100000, C367S140000, C367S135000

Reexamination Certificate

active

06173074

ABSTRACT:

FIELD OF THE INVENTION
This invention relates to acoustic signal analysis and in particular, to an acoustic signature recognition and identification system which alerts a user of machinery in proximity to a sensor placement.
BACKGROUND OF THE INVENTION
The measurement and analysis of vibrational and/or acoustic data is a widely used technique for determining the type or operating state of a machine. It is generally known that each type of machinery generates a plurality of frequencies dependent upon the operating condition of the machinery. For example, whether the machinery is idling, revving up or revving down changes the frequencies generated by the machinery. The changes in the frequency result from the interaction of the forces required to transition from one machine state to another. Importantly, these frequency shifts reflect changes in the characteristics of the machinery. The complex composite wave formed is unique for each type of machinery and is utilized for identification and recognition.
As illustrated in the prior art, the analysis of the acoustic data can be implemented in the time or frequency domain. In the frequency domain, the frequency spectrum of the machinery is termed an acoustic signature. By comparing the acoustic signature of the unknown or test machinery against a databank of collected acoustic signatures, it is possible to determine the state and/or type of machinery. In certain applications, it is important to determine both the type and state of the machinery. For example, in the telecommunications field thousands of telephone calls are carried over fiber optical cables which are buried underground. The owners can mitigate the consequences of any damage imposed by the machinery on the fiber optic cables by identifying the type and state of any machinery located near the fiber optic cables.
A drawback of the prior art devices is that they require human interaction to render a decision. Accordingly, there is a need to provide an acoustic signature identification and recognition system that alerts the owner by analyzing sound vibrations that impinge upon their equipment and by identifying the encroaching machinery.
SUMMARY OF THE INVENTION
The present invention teaches a system and method which utilizes the acoustic signatures of different types of machinery and their associated operating states, to provide alerts in response to specific conditions. This system utilizes minimal human intervention in rendering a decision. Importantly, it is adaptable to and applicable for numerous types of activities and interests, e.g., telecommunications cables and gas pipelines.
In an exemplary embodiment of the present method, acoustic signals received from an interferometer are converted to a digital format and are processed to create a spectrogram image of frequency versus time. The spectrogram image is then normalized to permit recognition of the acoustic pattern regardless of the magnitude of the acoustic signal. A feature extractor then detects, tracks and characterizes the lines which form the spectrogram. Specifically, the lines are detected via a KY process that is applied to each pixel in the line. A blob coloring process then groups spatially connected pixels into a single signal object. The harmonic content of the lines is then determined and compared with stored templates of known acoustic signatures for recognition and identification purposes.
Advantageously, the system and method of the present invention can be utilized to alert a party if any type of machinery is within the vicinity of, for example, a remotely located underground gas line, fiber optic cables for telephones or television or sewer lines. The above factors make the present invention a versatile system and method for machine recognition and identification.


REFERENCES:
patent: 3069654 (1962-12-01), Hough
patent: 5355208 (1994-10-01), Crawford et al.
patent: 5430690 (1995-07-01), Abel
patent: 5879303 (1999-03-01), Averkiou et al.
Ballard, D.H. and Brown, C.M.,Computer Vision, p. 151, Prentice-Hall, 1982.
Russo, A.P., “Constrained Neural Networks for Recognition of Passive Sonar Signals Using Shape,” IEEE Conference on Neural Networks in Ocean Engineering, Washinton, DC, pp. 69-76, Aug., 1991.

Affiliated with

Also associated with

Rating

Acoustic signature recognition and identification does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Acoustic signature recognition and identification, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Acoustic signature recognition and identification will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2534962