摘要
为了提高相关函数拟合极值法在二维位相板衍射光斑亚像素定位中的计算效率,对相关函数拟合极值法中的模板大小、拟合窗口大小及相关函数3种影响因素进行了分析。分析过程以仿真图像检验法为理论基础,即用数学仿真图像验证算法和程序的可靠性和精度。分析结果表明,3种影响因素均对计算效率产生较大影响,整像素搜索时模板和拟合窗口大小的选取对定位精度影响最大。同理,对二维位相板衍射光斑亚像素定位实验中的搜索模板和拟合窗口大小进行了优化设计,最佳模板大小为51 pixel×51 pixel,最佳拟合窗口大小为3 pixel×3 pixel,与选用恰好包含整个衍射光斑时的模板(131 pixel×131 pixel)相比,整个定位过程的计算时间缩短了40%,在不损失定位精度的同时有效提高了计算效率。
In order to improve the computational efficiency of sub-pixel location of a diffraction spot generated by a two-dimensional phase plate with the correlation function fitting extreme method, three effect factors, the template size, fitting window size and the correlation function were analyzed in detail. On the basis of the theory of emulation image examination, mathematical emulation images were used to verify the reliability and accuracy of algorithms and programs. Analyzing results indicate that the three effect factors all have bigger effect on computational efficiency and the template and fit- ting window sizes show the biggest effect on the locating precision during integer pixel searching. The optimal design of searching template and fitting window sizes are completed during the sub pixel loca tion experiment of diffraction spot generated by the two-dimensional phase plate,It is shown that the optimized template and fitting window sizes are 51 pixel× 51 pixel and 3 pixel× 3 pixel, respectively. Comparing with the template of 131 pixel× 131 pixel which just includes the whole diffraction spot, the calculating time of whole locating process is shortened by 40%, which shows the computational ef- ficiency is improved without degrading location precision.
出处
《光学精密工程》
EI
CAS
CSCD
北大核心
2009年第10期2392-2400,共9页
Optics and Precision Engineering
基金
Supported by the National Natural Science Foundation of China(GrantNo.60878028)
关键词
衍射光斑
计算效率
亚像素定位
相关函数
拟合窗口
diffraction spot
computational efficiency
sub-pixel location
correlation function
fitting window