CHOSUN

눈과 입 개폐의 상태검출을 통한 운전자의 졸음판단 분석

Metadata Downloads
Author(s)
공도현
Issued Date
2017
Abstract
In this paper, we propose an analysis method of drowsiness status of a driver from the opening and closing of eye and mouth. For this purpose, eyes, nose and mouth are detected to improve the area of interest using the
spatial correlation in the existing Viola-Jones algorithm. Then, histogram equalization is performed for detection at night driving, and drowsiness status of the driver data with the accumulation value through opening and closing of eye and mouth using SVM (Support Vector Machine) and PERCLOS (Percentage Closure of Eyes). The experimental result using Caltect face database showed that the detection rate of two eyes and nose were increased
by about 3% and 13.78%, respectively. Experimental results were confirmed that the performance of drowsiness through the opening and closing of the eyes was 90.72% and the performance of state through the opening and closing of the mouth was 98.26%.
Alternative Title
Analysis of Driver’s Drowsiness Through Status Detection from Opening and Closing of Eye and Mouth
Alternative Author(s)
Gong, Do Hyun
Affiliation
조선대학교
Department
일반대학원 제어계측공학과
Advisor
곽근창
Awarded Date
2017-08
Table Of Contents
제1장 서론 3
제2장 Viola-Jones 알고리즘을 이용한 운전자 얼굴 검출 방법 6
제1절 적분영상 7
제2절 Haar-like 특징 8
제3절 Adaboost 알고리즘 10 제4절 Cascade 구조 14
제3장 운전자 상태 및 졸음판단 방법 16
제1절 얼굴의 공간적 상관관계 16
제2절 운전자 눈과 입 개폐판단을 위한 과정 19
제3절 라벨링 방법 21
제4절 히스토그램 평활화 24
제5절 서포트 벡터 머신 26
1.선형 서포트 벡터 머신 분류 27
2.비선형 서포트 벡터 머신 분류 28
제6절 PERCLOS 29
제4장 실험 및 결과 31
제1절 데이터베이스 31
1.Caltech Face Data 31
2.운전자 주행 영상 32
제2절 유효성 판단 실험 34
제3절 입 개폐를 이용한 운전자 상태판단 36
제4절 눈 개폐를 통한 졸음판단 38
제5장 결론 44
참고문헌 45
Degree
Master
Publisher
조선대학교
Citation
공도현. (2017). 눈과 입 개폐의 상태검출을 통한 운전자의 졸음판단 분석.
Type
Dissertation
URI
https://oak.chosun.ac.kr/handle/2020.oak/13271
http://chosun.dcollection.net/common/orgView/200000266311
Appears in Collections:
General Graduate School > 3. Theses(Master)
Authorize & License
  • AuthorizeOpen
  • Embargo2017-08-25
Files in This Item:

Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.