摘要
工业CT是一种先进的无损检测手段。传统的滤波反投影算法(FBP)要求完备的投影数据,而实际检测中往往由于客观原因无法获得完全的投影数据。针对不完全投影的数据重建问题,提出了利用代数重建法(ART)进行图像重建。分析了影响ART算法的重建速度和重建质量的因素,通过仿真实验对两种算法的重建质量进行了比较。结果表明,在投影数据不完备时,ART算法的重建质量要优于FBP算法。
Industrial computed tomography（ICT） is an advanced means for nondestructive testing. Traditional filtered backprojection （FBP） method required complete projection data, which were often .difficult or even impossible to acquire for objective reasons. The algebraic reconstruction technique （ART） was proposed to reconstruct image from incomplete projection data. The factors influencing reconstruction speed and quality of ART were analyzed, the reconstruction quality of the two methods was compared by simulation experiments. The results showed that the reconstruction quality of ART was better than FBP when the projection data were incomplete.
出处
《无损检测》
北大核心
2007年第8期453-456,共4页
Nondestructive Testing
基金
国家自然科学基金资助项目(50375126)
航空基金资助项目(2006ZD53047)
陕西省教育厅科研基金资助项目(07JK425)
咸阳师范学院科研基金资助项目(06XSYK211)
关键词
工业CT
信号处理
图像重建
代数重建法
滤波反投影法
Industrial CT
Signal processing
Image reconstruction
Algebraic reconstruction technique
Filtered backprojection method
