Robust Image Projection System for the 3D Object based on Point Cloud Registration dealing with Point Sets having Different Numbers of Points

2015 Fiscal Year Final Research Report

• Project/Area Number: 25330181
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Perceptual information processing
• Research Institution: Hokkai-Gakuen University (2014-2015); Muroran Institute of Technology (2013)
• Principal Investigator: TAKAUJI Hidenori 北海学園大学, 工学部, 教授 (90431329)
• Co-Investigator(Kenkyū-buntansha): KANEKO Shun'ichi 北海道大学, 情報科学研究科, 教授 (50134789)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Keywords: 画像情報処理 / 視覚情報提示 / 3次元形状認識 / 点群照合 / 知能ロボティクス

Outline of Final Research Achievements

We developed the "2D projector" which can project a 2D image onto a 3D object and display images easily viewable for users.
Point cloud registration between the model point set and the data point set is used to obtain the detailed shape information of the 3D projection object. The model point set obtained from offline measurement is high-resolution and dense data. In contrast, the data point set measured by TOF camera is sparse data.
Additionally, we developed a distortion correction technique of projected images using the shape information of the 3D projection object obtained as point cloud data. The technique enables 2D image projection onto a curved concave-convex surface.

Free Research Field

情報学、知覚情報処理