페이지 마커를 이용한 문서 인식 및 증강
- Abstract
- Augmented reality (AR) have grown very popular because they provide users informative virtual contents in an intuitive way and interactivity to the virtual contents, thereby, attracting the interest of users. AR books, which is the main concern of this paper, is a representative one of AR applications. Thanks to their positive effects in education or entertainment, a number of AR books have been reported in the literature. However, almost all the existing AR books are available only on picture-based books or documents because the existing feature tracking methods are not useful for objects with poor texture such as text-based books or documents. Although use of fiducial markers or physical sensors is an easy way to implement AR on text-based books, the presence of markers reduces the users' immersion and the sensors require users to hold or wear additional physical equipment.
In this paper, we aim to implement AR on occluded text-based documents or books, especially distant ones where individual words or characters cannot be recognized. For this purpose, we propose a new paper-based AR system that converts documents or book pages to 2D binary markers, and identify them, estimate their relative 3D poses to the camera, and augment them with virtual contents. The system requires no fiducial markers or physical sensors. Experimental results demonstrate that the proposed system can recognize and augment occluded text-based documents or book pages with high accuracy and in real-time.
- Issued Date
- 2015
- Awarded Date
- 2015. 2
- Type
- Dissertation
- Publisher
- 부경대학교
- Affiliation
- 부경대학교 대학원 전자공학과
- Department
- 대학원 전자공학과
- Advisor
- 박한훈
- Table Of Contents
- Ⅰ 서 론 .............................................................................................................. 1
1.1 연구 배경 .................................................................................................... 1
1.2 연구 목적 및 필요성 ................................................................................... 2
Ⅱ 관련 이론 및 연구 ........................................................................................... 4
2.1 증강현실의 개념 .......................................................................................... 4
2.2 증강현실의 활용사례 .................................................................................... 6
2.2.1 방송, 광고 분야 ..................................................................................... 6
2.2.2 의료 분야 .............................................................................................. 6
2.2.3 산업 분야 .............................................................................................. 7
2.2.4 교육 분야 .............................................................................................. 7
2.3 증강현실의 구현 .......................................................................................... 9
2.3.1 마커 기반 증강현실 ............................................................................... 9
2.3.1.1 마커 검출 ........................................................................................ 10
2.3.1.2 마커의 종류 및 ID 인식 방법 ......................................................... 11
2.3.1.3 가상 콘텐츠 증강 ............................................................................ 14
2.3.2 비마커 기반 증강현실 ............................................................................ 20
2.3.2.1 에지 기반 ........................................................................................ 21
2.3.2.2 광류 기반 ........................................................................................ 22
2.3.2.3 텍스쳐 기반 .................................................................................... 23
2.3.2.4 비모델 기반 .................................................................................... 25
2.4 문서 인식 기술 ............................................................................................ 25
Ⅲ 문서 인식 및 증강 설계 ................................................................................... 29
3.1 시스템 개요 ................................................................................................. 29
3.2 문서 검출 .................................................................................................... 30
3.2.1 이진화 ................................................................................................... 30
3.2.2 문서 후보 영역 검출 ............................................................................. 32
3.3 페이지 마커 ................................................................................................. 39
3.4 페이지 마커 ID 분석 .................................................................................. 40
3.5 성능 개선 ..................................................................................................... 41
Ⅳ 실험결과 및 고찰 ............................................................................................. 44
4.1 실험 환경 ..................................................................................................... 44
4.2 실험 과정 ..................................................................................................... 44
4.3 결과 및 고찰 ................................................................................................ 49
4.4 추가 실험 ..................................................................................................... 53
4.4.1 서브영역 개수()에 따른 인식률 .......................................................... 53
4.4.2 인덱스(index)를 이용한 페이지 마커 ID 분석 ....................................... 53
4.4.3 특징점 기반 페이지 ID 분석 ................................................................. 55
4.4.4 그림 문서에 대한 실험 .......................................................................... 55
Ⅴ 결 론 ............................................................................................................... 60
Ⅵ 향후 계획 ......................................................................................................... 61
참 고 문 헌 ............................................................................................................ 62
- Degree
- Master
- Files in This Item:
-
-
Download
페이지 마커를 이용한 문서 인식 및 증강.pdf
기타 데이터 / 2.84 MB / Adobe PDF
-
Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.