一种应用于大范围视觉测量的二维卷积光重心法  

An Improved Optical Center of Gravity Method for Wide-Range Vision Measurement

作　　者：赵玲娜[1] 余有龙[1] 金兢[2] 李丹丹[1] 曹雪梅[1] 

机构地区：[1]合肥工业大学仪器科学与光电工程学院,安徽合肥230009 [2]中国科学技术大学工程科学学院,安徽合肥230001

出　　处：《光学与光电技术》2010年第5期76-79,共4页Optics & Optoelectronic Technology

摘　　要：机器视觉由于具有非接触、可视化好、自动化和智能性高等优点,在航空航天、国防、工业现场等重要领域得到广泛的应用,其特征点的检测精度会大大影响视觉测量的精度。在研究视觉测量的光点中心提取技术的过程中,提出二维卷积光重心法,可有效地抑制光点边缘对光点中心位置的影响,大大地提高了光点中心的提取精度,既可以应用于发光特征点的提取,也可以应用于反光特征点的提取。利用该方法提取空间特征点进行三维重构的精度也比其他方法要高,能达到10^(-4)左右,基本满足工程测量的要求。Vision measurement has a lot of advantages, such as non-contact, good visualization, automation and intelligence, so vision measurement has been widely used in a lot of important areas, like aerospace, national defense, industrial site. The detecting accuracy of laser feature points can affect the accuracy of vision measurement greatly. A 2D-convolution light center of gravity method for detecting the center of laser spots in vision measurement is proposed in this paper. This method can effectively reduce the influence of the laser spot fringe to the center position, which improves the detecting accuracy of laser spot center greatly. It can both apply to detecting light-emitting and reflective feature points. Using this method to detect spatial feature points can achieve three-dimensional reconstruction, and its accuracy is higher than other methods, which reachs about 10^-4. It satisfies the application in engineering measurement.

关 键 词：大范围 视觉测量 重心法 二维卷积 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]