Straight-Line-Trajectory-Based X-Ray Tomographic Imaging for Security Inspections: System Design, Image Reconstruction and Preliminary Results

被引:28
作者
Gao, Hewei [1 ]
Zhang, Li [1 ]
Chen, Zhiqiang [1 ]
Xing, Yuxiang [1 ]
Xue, Hui [1 ]
Cheng, Jianping [1 ]
机构
[1] Tsinghua Univ, Dept Engn Phys, Beijing 100084, Peoples R China
基金
中国国家自然科学基金;
关键词
Limited-angle reconstruction; linogram; straight-line rajectory; total variation; X-ray tomographic imaging; LIMITED-ANGLE TOMOGRAPHY; COMPUTED-TOMOGRAPHY; PROJECTION DATA; EXPLOSIVES DETECTION; ITERATIVE METHODS; CHECKED-BAGGAGE; BEAM CT; ALGORITHM; EXTRAPOLATION; REPROJECTION;
D O I
10.1109/TNS.2013.2274481
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
With the source and detector kept stationary while an inspected object moves between them, straight-line-trajectory-based tomographic imaging (SLTTI) has the capability to provide both digital radiologic images and tomographic images simultaneously, at high speed and with low-cost equipment. It has a promising future for fast security inspection for air cargo examinations. In order to fully investigate the feasibility of this approach, an experimental SLTTI platform has been developed. In this work, the system design is first presented. Then, a hybrid image reconstruction algorithm is proposed to obtain high quality tomographic images from the system to solve the limited-angle problem it represents. In the proposed algorithm, an initial reconstruction using either a direct filtered-backprojection (FBP)-type algorithm or the linogram technique is followed by a Gerchberg-Papoulis (GP) extrapolation to compensate for missing data. The GP extrapolation is implemented by the linogram technique, and constrained by a total variation (TV) minimization. By taking advantage of the FBP, the linogram and the TV techniques, we establish a fast and stable iterative reconstruction for the SLTTI. Finally, numerical simulations and realistic experiments are given, through which our proposals are validated to be feasible and effective. In the numerical examples, our algorithm can reduce the reconstruction inaccuracies for both high contrast and low contrast objects by factors of more than 15 and can generate nearly artifact free images as if complete projection data were available. A realistic experiment on a CatPhan (c) 600 phantom suggests that high quality tomographic images can be achieved from 120-deg of source fan-beam angle with undersampled data. For a group of explosive simulacra using 90 projection views over 120-deg of source fan-beam angle, the reconstruction inaccuracy can be reduced to lower than 4%.
引用
收藏
页码:3955 / 3968
页数:14
相关论文
共 51 条
[1]   ALGEBRAIC RECONSTRUCTION IN CT FROM LIMITED VIEWS [J].
ANDERSEN, AH .
IEEE TRANSACTIONS ON MEDICAL IMAGING, 1989, 8 (01) :50-55
[2]  
[Anonymous], 2001, Mathematical methods in image reconstruction
[3]  
[Anonymous], THESIS LINKOPING U L
[4]   Robust uncertainty principles:: Exact signal reconstruction from highly incomplete frequency information [J].
Candès, EJ ;
Romberg, J ;
Tao, T .
IEEE TRANSACTIONS ON INFORMATION THEORY, 2006, 52 (02) :489-509
[5]  
Chan T, 2006, HANDBOOK OF MATHEMATICAL MODELS IN COMPUTER VISION, P17, DOI 10.1007/0-387-28831-7_2
[6]   Prior image constrained compressed sensing (PICCS): A method to accurately reconstruct dynamic CT images from highly undersampled projection data sets [J].
Chen, Guang-Hong ;
Tang, Jie ;
Leng, Shuai .
MEDICAL PHYSICS, 2008, 35 (02) :660-663
[7]   Necessary and Sufficient Consistency Conditions for Fanbeam Projections Along a Line [J].
Clackdoyle, Rolf .
IEEE TRANSACTIONS ON NUCLEAR SCIENCE, 2013, 60 (03) :1560-1569
[8]   Globally convergent edge-preserving regularized reconstruction: An application to limited-angle tomography [J].
Delaney, AH ;
Bresler, Y .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 1998, 7 (02) :204-221
[9]   IMAGE-RECONSTRUCTION FROM LINOGRAMS - IMPLEMENTATION AND EVALUATION [J].
EDHOLM, P ;
HERMAN, GT ;
ROBERTS, DA .
IEEE TRANSACTIONS ON MEDICAL IMAGING, 1988, 7 (03) :239-246
[10]   LINOGRAMS IN IMAGE-RECONSTRUCTION FROM PROJECTIONS [J].
EDHOLM, PR ;
HERMAN, GT .
IEEE TRANSACTIONS ON MEDICAL IMAGING, 1987, 6 (04) :301-307