Depth Reconstruction of Nominally-Smooth Surfaces from Shadowed Images - Creativity Award

从阴影图像中深度重建标称光滑表面 - 创意奖

• 批准号: 9019951
• 负责人: Harold Vincent Poor
• 金额: $ 9万
• 依托单位: Princeton University
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1990
• 资助国家: 美国
• 起止时间: 1990-09-01 至 1994-02-28
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9019951&HistoricalAwards=false
• 关键词: Depth; Reconstruction; Nominally; Smooth; Surfaces

The extraction of three-dimensional information from two- dimensional views is an important aspect of the inspection of manufactured products. The determination of the quality of the product often depends on accurate three-dimensional information. Depth information is especially important in the inspection of nominally-smooth surfaces for defects. %%% This research intends to determine the three-dimensional structure of a nominally-smooth surface from a sequence of images acquired from a single point of view, where in each image, the object is illuminated by collimated light emanating from a unique direction.

从二维视图中提取三维信息是制造产品检测的一个重要方面。产品质量的确定往往依赖于准确的三维信息。深度信息在检查名义上光滑的表面是否有缺陷时尤其重要。本研究旨在从从单个视点获取的一系列图像中确定名义上平滑的表面的三维结构，其中在每一图像中，对象由从唯一方向发出的准直光照明。