基于最小特征根分析的手指基准点定位算法  

Location algorithm of finger fiducial point based on smallest eigenvalue analysis

作　　者：毛贤光[1] 赖晓铮[1] 赖声礼[1] 

机构地区：[1]华南理工大学电子与信息学院,广东广州510640

出　　处：《计算机工程与设计》2010年第7期1523-1525,1572,共4页Computer Engineering and Design

摘　　要：为准确定位手指基准点,提出了一种基于最小特征根分析的手指基准点定位算法。该方法利用曲线的非连续性特征,首先将手指轮廓构成的协方差矩阵以最小特征根的值来提取基准点区域,再采用拐点分析法矫正奇异区域,从而得到手指基准点。实验结果表明,该算法不仅能够减少计算量,还能有效克服噪声以及手形姿态变化的影响,并且对不同精度的图像都有较强的鲁棒性。To locate the finger fiducial point accuratly,a new location algorithm of finger fiducial point based on smallest eigenvalue analysis is presented.Based on uncontinuity of curve,this algorithm use single pixel and a curve which contain the single pixel as analysis object,and then utilize macroscopical collinearity to judge whether the target curve is the region of finger fiducial point or not.Owing to this,the smallest eigenvalue of covariance matrix get from these target pixels.Finnaly,through the critical point analyzing,the finger fiducial point is located exactly.Emulation experiment show that this algorithm not only reduce time-consuming,get rid of noise and hand-shape affection but also robust in different quality images.

关 键 词：曲线检测 协方差矩阵 最小特征根 手指基准点 定位 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]