Detection of Blurred Illumination Edges in the Structured Light 3D Scanning System

• Main Authors: Chuang-yi Chiou, 邱創義
• Other Authors: Cheng Wen
• Format: Others
• Language: zh-TW
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/17136878101838631441

碩士 === 龍華科技大學 === 電機工程研究所 === 96 === This paper proposes a blurred illumination edge model to determine the locations of illumination edges for the structured light 3D scanning systems. Differing from the conventional illumination edge detection techniques, which are mostly based on the gray level intensities, the proposed approach defines and utilizes the illumination ratio of each pixel on the captured image. Then an error-function based formulation against the proposed illumination ratio is derived to model a one-dimensional blurred illumination edge. In addition, the profile of the illumination ratios of a row of pixels is fitted with the proposed blurred illumination edge model to determine the locations of illumination edges. The major contribution of this paper is that the proposed illumination edge detector is accurate and robust, no matter the edges are in a wide or a narrow illumination region. Thus, this proposed technique not only works robustly, but also accelerates the measuring speed of the structured light 3D scanning systems.