Abnormal-sound detection device and abnormal-sound detection method

Patent metadata

Field	Value
Publication number	US-11971694-B2
Application number	US-201916964636-A
Country	US
Kind code	B2
Filing date	Feb 15, 2019
Priority date	Feb 16, 2018
Publication date	Apr 30, 2024
Grant date	Apr 30, 2024

Abstract

An abnormal-sound detection device has an imaging unit, an operation range identification unit, a sound collection unit, an abnormal-sound detection unit, an abnormal-sound generation position identification unit, and an abnormal-sound source determination unit. The operation range identification unit identifies and stores the operation range of a diagnosis object on the basis of the image captured by an imaging unit. The abnormal-sound detection unit detects abnormalities in sounds included in the sounds collected by the sound collection unit, the sounds arriving from the diagnosis object. When an abnormality in a sound is detected by the abnormal-sound detection unit, the abnormal-sound generation position identification unit identifies the position at which the abnormality of the sound was generated. The abnormal-sound source determination unit compares the operation range and the abnormal-sound generation position of the diagnosis object, and determines whether the abnormality of the sound is derived from an abnormality of the diagnosis object.

First claim

The invention claimed is:

1. An abnormal-sound detection device comprising:
at least one memory storing instructions; and
at least one processor configured to access the at least one memory and execute the instructions to:
extract a feature point of a diagnosis object from a video of the diagnosis object captured by a camera;
identify an operation range of the diagnosis object, based on a movement range of the extracted feature point;
sense a sound in which an abnormality has occurred from among sound collected by a microphone;
identify, based on the collected sound, a generation position of the sound in which the abnormality is sensed; and
determine, based on the operation range and the identified generation position, whether the sound in which the abnormality is sensed is derived from the diagnosis object or derived from another element.

2. The abnormal-sound detection device according to claim 1 , wherein
the at least one processor is further configured to execute the instructions to:
determine that the sound in which the abnormality is sensed is an abnormal sound of the diagnosis object, when the generation position of the sound in which an abnormality is sensed is in the operation range.

3. The abnormal-sound detection device according to claim 1 , wherein
the at least one processor is further configured to execute the instructions to:
determine that the sound in which the abnormality is sensed is an abnormal sound derived from a factor other than the diagnosis object, when the generation position of the sound in which an abnormality is sensed is out of the operation range.

4. The abnormal-sound detection device according to claim 1 , wherein
the at least one processor is further configured to execute the instructions to:
identify the generation position of the sound in which the abnormality is sensed, based on a difference of sounds each collected by each of microphones in a microphone array in which a plurality of microphones are arrayed.

5. The abnormal-sound detection device according to claim 1 , wherein
the at least one processor is further configured to execute the instructions to:
perform a machine learning to generate an abnormal characteristic determination criterion for determining an abnormal characteristic of the sound, based on sounds collected from the diagnosis object and a vicinity thereof.

6. An abnormal-sound detection method comprising:
capturing a video of a diagnosis object by a camera;
extracting a feature point of the diagnosis object from the video of the diagnosis object captured by the camera;
identifying an operation range of the diagnosis object, based on a movement range of the extracted feature point;
collecting sounds arriving from the diagnosis object by a microphone;
sensing a sound in which an abnormality has occurred from among the sounds collected by the microphone;
identifying, based on the collected sound, a generation position of the sound in which the abnormality is sensed; and
determining, based on the operation range and the identified generation position, whether the sound in which the abnormality is sensed is derived from the diagnosis object or derived from another element.

7. The abnormal-sound detection method according to claim 6 , further comprising
determining that the sound in which the abnormality is sensed is an abnormal sound of the diagnosis object, when the generation position of the sound in which an abnormality is sensed is within the operation range.

8. The abnormal-sound detection method according to claim 6 , further comprising
determining that the sound in which the abnormality is sensed is an abnormal sound derived from a factor other than the diagnosis object, when the generation position of the sound in which an abnormality is sensed is out of the operation range.

9. The abnormal-sound detection method according to claim 6 , further comprising:
collecting the sounds by use of a microphone array in which a plurality of microphones are arrayed; and
identifying the generation position of the sound in which the abnormality is sensed, based on a difference of the sounds each collected by each of the microphones.

10. A non-transitory computer-readable recording medium recording an abnormal-sound detection program that causes a computer to execute:
extracting a feature point of a diagnosis object from a video of the diagnosis object captured by a camera;
identifying an operation range of a diagnosis object, based on a movement range of the extracted feature point;
sensing a sound in which an abnormality has occurred from among sound collected by a microphone;
identifying, based on the collected sound, a generation position of the sound in which the abnormality is sensed; and
determining, based on the operation range and the identified generation position, whether the sound in which the abnormality is sensed is derived from the diagnosis object or derived from another element.

Assignees

• Nec Corp

Inventors

• Sendoda Mitsuru

Classifications

• G05B19/0425 Primary

  Safety, monitoring · CPC title

• G01S3/8083

  determining direction of source · CPC title

• G01S5/0063

  of measured values, i.e. measurement on base station and position calculation on mobile · CPC title

• G05B19/0428

  Safety, monitoring (G05B19/0423 takes precedence) · CPC title

• G06N20/00

  Machine learning · CPC title