Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Kwak, Su Ha | - |
dc.contributor.author | Park, Jin Hee | - |
dc.date.accessioned | 2017-08-02T07:33:38Z | - |
dc.date.available | 2017-08-02T07:33:38Z | - |
dc.date.issued | 2017 | - |
dc.identifier.uri | http://dgist.dcollection.net/jsp/common/DcLoOrgPer.jsp?sItemId=000002376733 | en_US |
dc.identifier.uri | http://hdl.handle.net/20.500.11750/4104 | - |
dc.description.abstract | Accurate classification of human aquatic activities using radar has a variety of potential applications such as rescue operations and border patrols. Nevertheless, the classification of activities of water using radar has not been extensively studied, unlike the case of dry ground, due to its unique challenge. Namely, not only is the radar cross section of a human on water small, but the micro-Doppler signatures are much noisier due to water drops and waves. In this paper, we first investigate whether discriminative signatures could be obtained for activities on water through a simulation study. Then we show how we can effectively achieve high classification accuracy by applying deep convolutional neural networks(DCNN) directly to the spectrogram of real measurement data. From the five-fold cross-validation on our dataset, which consists of five aquatic activities, we report that the conventional feature-based scheme only achieves an accuracy of 45.1%. In contrast, the DCNN trained using only the collected data attains 66.7%, and the transfer learned DCNN, which takes a DCNN pre-trained on a RGB image dataset and fine-tunes the parameters using the collected data, achieves a much higher 80.3%, which is a significant performance boost. ⓒ 2017 DGIST |
- |
dc.description.statementofresponsibility | open | - |
dc.description.tableofcontents | Ⅰ. Introduction 1-- 1.1 Overview 1-- 1.2 Related work 3-- 1.2.1 Human Activity Classification on Ground using Convolutional Neural Networks 3-- 1.2.2 Human Activity Classification on Water using Convolutional Neural Networks 4-- Ⅱ. Micro-Doppler Simulation of Swimming Activities 5-- Ⅲ. Measurements of Human Activities on Water 7-- Ⅳ. Classification Method Using Deep Convolutional Neural Networks 9-- 4.1. DCNN Trained from Scratch 9-- 4.2. Transfer Learned DCNN 10-- Ⅴ. Experimental Results 13-- Ⅵ. Discussion and Conclusions 17 |
- |
dc.format.extent | 23 | - |
dc.language | eng | - |
dc.publisher | DGIST | - |
dc.subject | radar | - |
dc.subject | micro-Doppler signatures | - |
dc.subject | aquatic activity classification | - |
dc.subject | 레이더 | - |
dc.subject | 마이크로 도플러 신호 | - |
dc.subject | 수중행동분류 | - |
dc.subject | 전이학습 | - |
dc.subject | 컨볼루션 신경망 | - |
dc.title | Micro-Doppler Based Classification of Human Activities on Water via Transfer Learning of Convolutional Neural Networks | - |
dc.title.alternative | CNN 전이학습을 이용한 마이크로 도플러 신호 기반의 사람 수중행동 분류 | - |
dc.type | Thesis | - |
dc.identifier.doi | 10.22677/thesis.2376733 | - |
dc.description.degree | Master | - |
dc.contributor.department | Information and Communication Engineering | - |
dc.contributor.coadvisor | Moon, Tae Sup | - |
dc.date.awarded | 2017. 8 | - |
dc.publisher.location | Daegu | - |
dc.description.database | dCollection | - |
dc.date.accepted | 2017-07-31 | - |
dc.contributor.alternativeDepartment | 대학원 정보통신융합공학전공 | - |
dc.contributor.affiliatedAuthor | Kwak, Su Ha | - |
dc.contributor.alternativeName | 박진희 | - |
dc.contributor.alternativeName | 곽수하 | - |
dc.contributor.alternativeName | 문태섭 | - |