개인구독
소속 기관이 없으신 경우, 개인 정기구독을 하시면 저렴하게
논문을 무제한 열람 이용할 수 있어요.
지원사업
학술연구/단체지원/교육 등 연구자 활동을 지속하도록 DBpia가 지원하고 있어요.
커뮤니티
연구자들이 자신의 연구와 전문성을 널리 알리고, 새로운 협력의 기회를 만들 수 있는 네트워킹 공간이에요.
논문 기본 정보
- 자료유형
- 학위논문
- 저자정보
- 지도교수
- 안현식
- 발행연도
- 2016
- 저작권
- 동명대학교 논문은 저작권에 의해 보호받습니다.
이용수2
이 논문의 연구 히스토리 (2)
- 이 논문의 후속연구가 궁금하신가요?
- 연관 학술논문 또는 학술발표를 통해 보다 발전된 연구결과를 확인하실 수 있습니다.
- 이 논문의 연구 히스토리 확인하기
초록· 키워드
오류제보하기
본 논문에서는 인지 로봇에 있어서 물체를 인식하는 과정과 이를 토대로 문장기반 인지시스템을 배경으로 하여 3D 물체를 인지하고 언어와 연결된 물체와 동작으로 인간의 명령에 따라 동작하는 인지로봇을 연구한다. 먼저 3D 물체의 인식을 위해 깊이검출 카메라로부터 RGB-D 데이터를 입력하고 3D 데이터의 노이즈 저감과 영역분할을 연구한다. 깊이검출 카메라로부터 입력된 깊이 정보는 노이즈를 포함하고 있는데, 이러한 노이즈를 제거하기 위하여 법선성분을 이용한 3D 메디언(Median) 필터를 활용한다. 노이즈가 제거된 각 점의 법선벡터로 부터 이웃하는 점들의 법선벡터들과의 내적을 취함으로서 벡터의 방향 변화율 정도를 계산하여 구분하고 유사한 특성을 나타내는 위치 점을 분할하기 위해 영역성장(region growing) 기법을 적용한다. 깊이검출 카메라는 일반적으로 한 시점에서의 거리 데이터를 검출하게 된다. 이러한 검출 데이터는 물체의 한 단면만을 나타내므로 다른 시점에서 검출된 거리 데이터와의 통합 즉 일반화(generalization)가 필요하다. 본 논문에서는 컬러 영상 내에 포함된 특징을 이용하여 물체에 대한 정합을 한 후 이를 이용해 두 데이터를 일반화하는 방법을 제시한다.
인식된 물체는 문장기반 인지시스템의 시각모듈에서 사건으로 표현되고 이는 사건해석기에서 문장으로 표현되며 동시에 물체기술자(object descriptor)에 대상 물체의 3D 데이터가 저장된다. 이러한 과정에서 물체는 지표와 탄도체의 상호 관계로 표현되고 인지적 개념을 언어적 표현으로 나타낸다. 또한 물체를 처리하는 동작을 동작 기술자의 계층적 모델을 기반으로 하여 정의하고 물체에 대한 처리를 함과 동시에 그 동작 사건을 문장으로 생성하여 인지시스템의 문장메모리에 저장한다. 이러한 과정을 통해 물체에 대한 3D 인지와 이에 대한 인지시스템 기반의 물체처리 방법론을 제시한다.
인식된 물체는 문장기반 인지시스템의 시각모듈에서 사건으로 표현되고 이는 사건해석기에서 문장으로 표현되며 동시에 물체기술자(object descriptor)에 대상 물체의 3D 데이터가 저장된다. 이러한 과정에서 물체는 지표와 탄도체의 상호 관계로 표현되고 인지적 개념을 언어적 표현으로 나타낸다. 또한 물체를 처리하는 동작을 동작 기술자의 계층적 모델을 기반으로 하여 정의하고 물체에 대한 처리를 함과 동시에 그 동작 사건을 문장으로 생성하여 인지시스템의 문장메모리에 저장한다. 이러한 과정을 통해 물체에 대한 3D 인지와 이에 대한 인지시스템 기반의 물체처리 방법론을 제시한다.
목차
Ⅰ. 서 론·············································································································11. 연구의 목적··································································································12. 기존의 연구··································································································23. 논문의 구성··································································································4Ⅱ. 인지로봇의 3D 물체의 인지··································································61. RGB-D 데이터 입력 및 3D 데이터 표현··············································62. 3D 물체 데이터 필터링···········································································113. 3D 물체 데이터 영역분할 ·······································································144. 3D 물체 데이터의 일반화 ·······································································16Ⅲ. 문장기반 인지시스템을 이용한 물체의 처리·································201. 물체 처리를 위한 로봇 시스템· ·····························································202. 문장기반의 인지시스템 ············································································223. 3D 물체의 인지 언어적 표현·································································244. 문장기반 인지시스템의 물체의 처리···················································· 26Ⅳ. 실험 및 고찰······························································································291. 실험용 시스템 구성 ··················································································292. 실험 및 결과 ······························································································30Ⅴ. 결론···············································································································461. 논문요약·····································································································462. 추후 연구 방향························································································47참고문헌 ···········································································································48ABSTRACT· ··································································································52
최근 본 자료
댓글(0)
0