摘要
为了解决快中子编码成像中复杂源图像定义与模拟效率的问题,研究了基于深度优先合并方法的复杂源图像定义方法。采用栅元合并技术来减小源定义所需栅元数目,同时保证蒙特卡洛模拟中源抽样的效率,从而提高复杂源图像编码成像模拟效率。模拟源为二值E字母时,源合并后模拟所得编码像的计算结果最小误差较源合并前没有变化且满足统计要求,计算时间则减少且为源合并前模拟时间的1/5;对16、64和256灰度阶E字母源进行了编码像的模拟计算,模拟结果的最小误差小于1%,符合重建研究的需要;采用3种重建算法对"西安交通大学校徽"复杂二值源的定义和模拟进行了源区重建,进而验证了基于深度优先合并的源定义方法的正确性。该方法可望为增进聚变源区所历复杂过程诊断的适应性提供一种切实可行的技术途径。
To solve the difficulties in complex gray-scale source definition and simulation efficiency,a strategy for defining complex gray-scale source based on the depth-first merger is proposed.A technology of cells merging is used to reduce the number of cells for the source definition and ensure the sampling efficiency of source in Monte Carlo simulation,thus the simulation efficiency of coded imaging of the complex gray-scale source can be increased.When the simulated source is the binary letter E source,the error of simulated coded image of the merged source remains same as that of the source without merging and meets the requirement of the reconstruction for statistics,while the computing time for merged source is reduced to 1/5of the source without merging.The 16,64 and 256gray-scale letter E sources are defined by the definition method of complex gray-scale source based on the depth-first merger,and the corresponding simulations of coded image are also carried out.The errors of simulation results reach less than 1%,which meet the requirement of the reconstruction for statistics.The complex binary source of Xi'an Jiaotong University school badge is defined with the proposed strategy forcomplex gray-scale source based on the depth-first merger and the corresponding simulation of coded image is carried out,then the reconstructions of the source by three different methods are realized.The strategy for defining complex gray-scale source based on the depth-first merger is verified,so it can be expected to provide a feasible technological approach to enhance the adaptability of diagnosis of the complex capsule implosion process.
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2016年第7期118-124,共7页
Journal of Xi'an Jiaotong University
基金
教育部高校创新研究团队资助项目(IRT1280)
陕西省自然科学基础研究计划资助项目(2015JZ001)
国家自然科学基金资助项目(11505166)
关键词
快中子编码成像
深度优先合并
复杂灰度源
蒙特卡洛模拟
图像重建
fast neutron coded imaging
depth-first merger
complex gray-scale source
Monte Carlo simulation
image reconstruction