MI 센서기반의 금속탐지를 위한 RNN 알고리즘 설계에 관한연구

• 영문명: A Study on the RNN Algorithm for Metal Detection Based on MI Sensor
• 발행기관: 한국산업기술융합학회(구. 산업기술교육훈련학회)
• 저자명: 하성재(Sungjae Ha) 이동우(Dongwoo Lee) 김회준(Hoijun Kim) 김응조(Eung-Jo Kim) 권순철(Soonchul Kwon) 이승현(Seunghyun Lee)
• 간행물 정보: 『산업기술연구논문지』산업기술연구논문지 제26권 2호, 103~111쪽, 전체 9쪽
• 주제분류: 공학 > 공학일반
• 파일형태: PDF
• 발행일자: 2021.06.30

영문 초록

This paper is a study of metal detection methods using multiple MI sensors using deep learning. The MI sensor is a sensor that measures the impedance change of an atypical wire by the electrical conduction and transmittance of a detector. Simple principles, small power consumption, and compact manufacturing are possible. This is a field-available sensor that requires limited weight and power, such as a mobile environment. However, sensors are generally weak in strength due to the nature of sensors that detect changes in magnetic fields. This paper compares and analyzes the detection performance with the depth of each network for location detection via RNN on 16-channel raw data extracted from sensors in a self-impedance manner.

목차

Ⅰ. 서 론

Ⅱ. 관련연구

Ⅲ. 네트워크

Ⅳ. 실험결과

Ⅴ. 결 론

키워드

딥러닝 전자기유도 자기임피던스센서 다중센서 순환신경망 Deep Learning Electromagnetic Induction MI Sensor Multi Sensor RNN