基于γ图像重建放射源三维几何形状的模拟研究  

Simulation Research in 3D Reconstruction of Radioactive Sources' Geometrical Shapes by Gamma Images

作　　者：朱宁[1] 党晓军[1] 兰明聪[1] 张连平[1] 吴伦强[1] 韦孟伏[1] 

机构地区：[1]中国工程物理研究院,绵阳621900

出　　处：《核电子学与探测技术》2015年第3期239-243,共5页Nuclear Electronics & Detection Technology

摘　　要：准确评估待检测对象内放射性物质的几何形状,在核设施退役和核安全等方面都具有重要的意义。利用MCNP模拟屏蔽体内不同几何形状的体源,针孔γ相机围绕被探测区域获取投影数据。根据针孔成像的机理,建立系数矩阵,并通过最大似然期望最大(MLEM)算法重建体源的三维位置和形状分布。模拟数据结果表明:该方法可准确地重建出放射性体源的三维位置和形状分布,体源在三维方向上形状的尺寸偏差不大于10%。To define the radioactive sources＇ geometrical shapes in measured objects is significant in the areas of nuclear waste and nuclear facilities. The radioactive volume sources with different appearance in attenuation medium and pinhole gamma eamera surrounding the deteeted space to get projections are simulated by MCNP. Based on the principle of pinhole imaging, the coefficient matrix is set up and the 3D distribution of volume sources＇ positions and shapes are reconstructed by the maximization likelihood expectation maximization （MLEM） algorithm. The experimental results show that this method earl reconstruct the positions and shapes of volume sources accurately and the deviation of shape in a 3D direction is not more than 10%.

关 键 词：Γ相机 针孔成像 最大似然期望最大(MLEM)算法 

分 类 号：TP391.75[自动化与计算机技术—计算机应用技术]