摘要
现有的工业CT重建算法均为全局物体的断层成像,即使只需要重建物体中的一个相当小的区域,也必须通过扫描整个物体的断层来获取投影数据,并利用投影数据重建整个物体,利用迭代法重建数据量大且耗费时间长,难以满足工程需要.本文拟先对整个固体火箭发动机工件进行低分辨率的初次快速解析法重建,可视化选取感兴趣区域,再对感兴趣区域进行迭代法精细高分辨率重建,实现工业CT缺陷区域的快速精细重建.
All the existing industrial CT reconstruction algorithm are the tomography of global object, even if you only need to reconstruct a relatively small area of the object, you have to obtain the projec- tion data by scanning the fault of the whole object and use the projection data to reconstruct the whole object. The data volume is large and it takes a long time by using the iterative methods to reconstruct and ifs hard to meet the needs of engineering. The paper intended to have a rapid and low resolution to the entire solid rocket engine, visually select the region of interested, finally use iterative method of high resolution reconstruction to rebuild the region of interested, the method realize the rapid and precisely reconstruction for the defect area in industrial CT.
出处
《测试技术学报》
2017年第6期537-540,共4页
Journal of Test and Measurement Technology
基金
国家自然基金资助项目(61171177)
国家重大科学仪器设备开发专项资助课题(2013YQ240830)
山西省科技攻关资助项目(20140321010-02)
关键词
无损检测
感兴趣区域
图像重建
迭代法
固体火箭发动机
nondestructive testing
interested region
image reconstruction
ART
solid rocket engine
