Anomaly detection apparatus, method, and program

Patent metadata

Field	Value
Publication number	US-2021256312-A1
Application number	US-201817056070-A
Country	US
Kind code	A1
Filing date	May 18, 2018
Priority date	May 18, 2018
Publication date	Aug 19, 2021
Grant date	—

Abstract

An anomaly detection apparatus comprises a pattern storage part, first long time-span feature extraction part, a pattern feature calculation part, and a score calculation part. The pattern storage part stores a signal pattern model trained based on an acoustic signal for training in a first time-span and a feature for training being calculated from signal for training in a second time-span that is longer than first time-span. The first long time-span feature extraction part extracts a long time-span feature for detection associated with the feature for training from a signal being a detection target. The pattern feature calculation part calculates a signal pattern feature related to a signal being a detection target based on the signal being a detection target, the feature for detection, and the signal pattern model. The score calculation part calculates score to detect anomaly in the signal being detection target.

First claim

1 . An anomaly detection apparatus, comprising:
a memory configured to store instructions, and a processor configured to execute the instructions, the instructions comprising: storing a signal pattern model trained based on an acoustic signal for training in a first time-span, and a long time feature for training calculated from an acoustic signal for training in a second time-span that is longer than the first time-span; extracting a long time-span feature for anomaly detection associated with the long time feature for training from an acoustic signal being a target of anomaly detection; calculating a signal pattern feature related to an acoustic signal being a target of anomaly detection based on the acoustic signal being a target of anomaly detection, the long time feature for anomaly detection and the signal pattern model; and a score calculating an anomaly score to detect anomaly in the acoustic signal being a target of anomaly detection, based on the signal pattern model.

2 . The anomaly detection apparatus according to claim 1 , wherein the instructions further comprise:
buffering the acoustic signal for anomaly detection during at least the second time-span.

3 . The anomaly detection apparatus according to claim 2 , wherein the instructions further comprise:
extracting an acoustic feature based on the acoustic signal for anomaly detection that is outputted from the buffer part, and extracting the long time-span feature for anomaly detection based on the acoustic feature.

4 . The anomaly detection apparatus according to claim 1 , wherein the signal pattern model is a prediction device that estimates a probability distribution to be followed by the acoustic signal being a target of the anomaly detection at time t+1 by receiving an input of the acoustic signal being a target of the anomaly detection at time t.

5 . The anomaly detection apparatus according to claim 4 , wherein the signal pattern feature is expressed as a series of probability values for each possible value taken by the acoustic signal being a target of anomaly detection at time t+1, and
the score calculating an entropy of the signal pattern feature, to calculate the anomaly score using the calculated entropy.

6 . The anomaly detection apparatus according to claim 1 , wherein the instructions further comprise:
storing a long time-span signal model at least as a reference to extract the long time-span feature for anomaly detection, wherein the extracting extracts the long time-span feature with further reference to the long time-span signal model for anomaly detection.

7 . The anomaly detection apparatus according to claim 1 , wherein the acoustic signal for training and the acoustic signal for anomaly detection are acoustic signals generated by a generating mechanism providing a change of state.

8 . The anomaly detection apparatus according to claim 1 , wherein the instructions further comprise:
extracting the long-time span feature for training, and performing training of the signal pattern model based on the acoustic signal for training and the long time-span feature for training.

9 . An anomaly detection method, in an anomaly detection apparatus that comprises a pattern storage part that stores a signal pattern model trained based on an acoustic signal for training in a first time-span, and a long time feature for training calculated from an acoustic signal for training in a second time-span that is longer than the first time-span,
the method comprising: extracting a long time-span feature for anomaly detection associated with the long time feature for training from an acoustic signal being a target of anomaly detection; calculating a signal pattern feature related to an acoustic signal being a target of anomaly detection based on the acoustic signal being a target of anomaly detection, the long time feature for anomaly detection and the signal pattern model; and calculating an anomaly score to detect anomaly in the acoustic signal being a target of anomaly detection, based on the signal pattern model.

10 . A computer-readable recording medium storing a program for causing a computer installed in an anomaly detection apparatus that comprises a pattern storage part that stores a signal pattern model trained based on an acoustic signal for training in first time-span, and a long time feature for training calculated from an acoustic signal for training in a second time-span that is longer than the first time-span, to execute:
a process of extracting a long time-span feature for anomaly detection associated with the long time feature for training from an acoustic signal being a target of anomaly detection; a process of calculating a signal pattern feature related to an acoustic signal being a target of anomaly detection based on the acoustic signal being a target of anomaly detection, the long time feature for anomaly detection and the signal pattern model; and a process of calculating an anomaly score to detect anomaly in the acoustic signal being a target of anomaly detecting, based on the signal pattern model.

11 . The anomaly detection method according to claim 9 , further comprising:
buffering the acoustic signal for anomaly detection during at least the second time-span.

12 . The anomaly detection method according to claim 11 , further comprising:
extracting an acoustic feature based on the acoustic signal for anomaly detection that is outputted from the buffer part, extracting the long time-span feature for anomaly detection based on the acoustic feature.

13 . The anomaly detection method according to claim 9 , wherein the signal pattern model is a prediction device that estimates a probability distribution to be followed by the acoustic signal being a target of the anomaly detection at time t+1 by receiving an input of the acoustic signal being a target of the anomaly detection at time t.

14 . The anomaly detection method according to claim 13 , wherein the signal pattern feature is expressed as a series of probability values for each possible value taken by the acoustic signal being a target of anomaly detection at time t+1, and
the score calculating an entropy of the signal pattern feature, to calculate the anomaly score using the calculated entropy.

15 . The anomaly detection method according to claim 9 , further comprising:
storing a long time-span signal model at least as a reference to extract the long time-span feature for anomaly detection, wherein the extracting extracts the long time-span feature with further reference to the long time-span signal model for anomaly detection.

16 . The anomaly detection method according to claim 9 , wherein the acoustic signal for training and the acoustic signal for anomaly detection are acoustic signals generated by a generating mechanism providing a change of state.

17 . The anomaly detection method according to claim 9 , further comprising:
extracting the long-time span feature for training, and performing training of the signal pattern model based on the acoustic signal for training and the long time-span feature for training.

18 . The medium according to claim 10 , the program further comprising:
buffering the acoustic signal for anomaly detection during at least the second time-span.

19 . The medium according to claim 18 , the program further comprising:
extracting an acoustic feature based on the acoustic signal for anomaly detection that is outputted from the buffer part, and extracting the long time-span feature for anomaly detection base

Assignees

• Nec Corp
• National Univ Corporation Tokai National Higher Education And Research System

Inventors

• Komatsu Tatsuya
• Kondo Reishi
• Hayashi Tomoki

Classifications

• G10L25/51 Primary

  for comparison or discrimination · CPC title

• G06F18/2148

  characterised by the process organisation or structure, e.g. boosting cascade · CPC title

• G06F18/241

  relating to the classification model, e.g. parametric or non-parametric approaches · CPC title

• G06N3/042

  Knowledge-based neural networks; Logical representations of neural networks · CPC title

• G06N3/0442

  characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] · CPC title