Invention Grant
- Patent Title: Anomaly detection system and method using noise signal and adversarial neural network
-
Application No.: US17469514Application Date: 2021-09-08
-
Publication No.: US11869492B2Publication Date: 2024-01-09
- Inventor: Harsh Pramodbhai Purohit , Takashi Endo , Yohei Kawaguchi
- Applicant: HITACHI, LTD.
- Applicant Address: JP Tokyo
- Assignee: HITACHI, LTD.
- Current Assignee: HITACHI, LTD.
- Current Assignee Address: JP Tokyo
- Agency: Volpe Koenig
- Priority: JP 20216561 2020.12.25
- Main IPC: G05B23/02
- IPC: G05B23/02 ; G10L15/20 ; G10L15/02 ; G06N3/08 ; G10L17/18 ; G10L25/78

Abstract:
An anomaly detection apparatus includes a device identification database that stores device identification information for identifying a specific device for each type of a device, a hierarchical conditional vector generation unit that generates a hierarchical conditional vector based on the device identification information, an extraction unit that extracts a target device feature amount vector indicating a feature amount of an acoustic signal acquired from a target device by analyzing the acoustic signal, a hierarchical condition adversarial neural network that outputs background noise level information indicating a background noise level of a surrounding environment of the target device and true/false determination information indicating true/false of the target device feature amount vector by analyzing the hierarchical conditional vector and the target device feature amount vector, and an anomaly determination unit that determines whether an anomaly exists in the target device feature amount vector.
Public/Granted literature
- US20220208184A1 ANOMALY DETECTION APPARATUS, ANOMALY DETECTION METHOD, AND ANOMALY DETECTION SYSTEM Public/Granted day:2022-06-30
Information query