摘要
针对复杂高原子序数物品在安检中难探测的问题,提出通过X射线安检图像两次灰度变换的算法,分别将图像中复杂高原子序数物品整体和内部进行分割:首先利用双能X射线成像原理得到物质材料值图像;然后根据高原子序数物品材料值特征对得到的材料值图像压缩进行类伽马变换;再利用新材料值信息将高原子序数区域分割出来;最后对高原子序数区域图像拉升进行第二次伽马变换,利用K均值分类将内部重叠区域分割出来。
Aiming at the problem that complex high atomic number objects are difficult to detect in security inspection,this paper proposes the algorithm of twice gray transformation of X-ray image to separate the whole and the interior of complex high-atomic number objects in the image respectively：Firstly,the material image is obtained by using the principle of dual-energy X-ray imaging. Then,the image data of the material is subjected to Gamma-like transformation based on the value of the high atomic number material. Then,the high atomic number region is segmented by using the new material value information. At last,the second gamma transformation is performed on the image of the high atomic number region,and the inner overlapping region is segmented by K-means classification.
作者
白聪
白廷柱
查艳丽
张硕
BAI Cong;BAI Ting-zhu;ZHA Yan-li;ZHANG Shuo(Beijing Institute of Technology, Beijing 100081, China;Ministry of Public Security First Institute, Beijing 102200, China)
出处
《核电子学与探测技术》
北大核心
2017年第10期963-968,共6页
Nuclear Electronics & Detection Technology
关键词
安检图像
图像分割
灰度变换
重叠
Security image
Image segmentation
Gray transform
overlapping
