结构光光条中心点信度评价方法与应用  

Reliability Evaluation Method and Application for Light-Stripe-Center Extraction

作　　者：王顺[1] 徐静珠[1] 张益昕[1] 张旭苹[1] 谢飞[1] 

机构地区：[1]南京大学光通信工程研究中心,江苏南京210008

出　　处：《光学学报》2011年第11期203-209,共7页Acta Optica Sinica

基　　金：江苏省自然科学基金面上项目(BK2009257);江苏省2010年度普通高校研究生科研创新基金(CX10B_020Z)资助课题

摘　　要：结构光三维测量中,光条中心点的提取是极其重要的步骤。现有方法对结构光条中心点提取都仅仅是标志中心点的位置,而中心点位置的可信与否,并未提及。对于一些特殊场合应用,例如室外复杂情况的测量,需要确定测量结果的可信程度。如何评价光条提取的中心点的可信程度成为亟待解决的关键问题。针对这一问题,对光条模型进行实际分析,给出了更为合理的高斯模型为光条模型。在此基础上以光条截面归一化高斯模型能量和为信度评价依据,提出了结构光条中心点信度评价方案,同时进行了实验验证。结果表明,此评价结果可以有效地评价区分各种情况的光条中心点的可信程度,更加符合实际评价需求。In line structured light vision measurement, extraction ot the llgiat-strlpe center is an extremely important step. The existing extraction methods simply give out the stripe center position. Few take it into account how reliable of the extraction result is. In some special occasions, such as outdoor measurement, the reliability of the measurement results is greatly needed. Therefore, how to evaluate the reliability of the stripe center becomes a key problem. A reliability evaluation method for light-stripe-center extraction has been proposed, which is on the basis of the total energy of stripe cross section normalized by Gaussian model. Experimental results show that this method can effectively evaluate stripe center reliability in different situations, and the result is more consistent with practical need.

关 键 词：机器视觉 结构光条 高斯模型 信度评价 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]