公开(公告)号：US20210216851A1

公开(公告)日：2021-07-15

申请号：US17145289

申请日：2021-01-09

Applicant: Electronics and Telecommunications Research Institute

Inventor： Byung Bog LEE ,  Woong Shik YOU ,  Gi Young LEE ,  Cheol Sig PYO

IPC: G06N3/04 ,  G06N3/08

Abstract: According to one embodiment disclosed herein, an edge device may include: a sensor circuit capable of sensing a surrounding environment; and a processor, wherein the processor is configured to: obtain sensor data about the surrounding environment through the sensor circuit; identify a notification urgency of the sensor data by performing an inference based on a neural network graph via an artificial intelligence inference engine; and adjust a notification period of the sensor data depending on the identified urgency.

公开(公告)号：US20210271957A1

公开(公告)日：2021-09-02

申请号：US17174199

申请日：2021-02-11

Applicant: Electronics and Telecommunications Research Institute

Inventor： Gi Young LEE ,  Byung Bog LEE ,  Woong Shik YOU ,  Cheol Sig PYO

IPC: G06N3/04 ,  G06N3/08 ,  G06N20/00 ,  G06N5/04

Abstract: The invention relates to a technology for detecting an abnormal signal using a filter for removing normal sound (or normal signals) around a sensor at normal times. The filter is provided to remove normal sound based on a denoising autoencoder learning technique for removing noise and used to determine whether field sound is an abnormal signal different from that of normal times. The filter is trained to pass normal sound, regarded as noise, to output a value of 0 and pass an abnormal signal without change. The filter is retrained by collecting only normal sound rather than abnormal signals in the field and then adding the collected normal sound to the existing training data. Therefore, even machine-learning nonexperts may easily and conveniently retrain the filter.