摘要
针对数字散斑图像的极值位移测量法存在的极值点对误匹配问题,提出一种改进的整像素极值位移法。首先通过极值法查找移动前后散斑图的极值点,构造极值矩阵,生成三维显示图;然后利用多个指定灰度平面平截三维显示图,得到平截点;最后分析平截点构成的相对位移矩阵,计算物体位移。在无噪声和有噪声的条件下进行仿真实验,实验结果都证明,改进算法在保证位移测量精度准确的条件下,极值点对误匹配数减少,运算效率提高103倍。将该算法应用到激光鼠标定位中,移动位移分辨力可达1μm,移动方向的角度误差不超过2.72°。结论证明,融合多层平截的激光散斑图像极值位移法是一种快速、高效、实用的改进算法。
Aiming at the problem of mismatching of the extreme points in extremum method for digital speckle images, a new algorithm named the whole pixel extremum displacement measurement for laser speckle images with multi-level crossing was proposed. The extremum method was used to figure out the extreme points of speckle images before and after displacement, constructed extremum value matrix and generated 3-D figure. Then the truncation points were got by multiple specified gray planes crossing 3-D figure. And relative displacement matrix of truncation points was analyzed to calculate displacement of object. The simulation results under the condition of no noise and noise prove that the improved algorithm under guaranteeing the accuracy of displacement measurement, decreases the number of extreme points of error matching and improves operation efficiency 10^3 times. The algorithm was applied to the laser mouse positioning and the experimental results show that the displacement of the mobile resolution reached 1 microns, and moving direction angle error was less than 2. 72°.It is conclued proves that multi-level truncation extremum displacement method based on laser speckle image is a rapid,efficient and practical algorithm.
出处
《计算机应用》
CSCD
北大核心
2015年第5期1430-1434,1443,共6页
journal of Computer Applications
基金
总装备部预研基金资助项目
关键词
数字散斑图像
极值位移测量法
多层平截
极值点
平截点
digital speckle image
Extremum Displacement Measurement （EDM）
multi-level crossing
extreme point
truncation point
