2021 Fiscal Year Final Research Report
Demonstration of translated X-ray CT using reconstruction algorithm based on mathematical optimization method
Project/Area Number |
19K15476
|
Research Category |
Grant-in-Aid for Early-Career Scientists
|
Allocation Type | Multi-year Fund |
Review Section |
Basic Section 31010:Nuclear engineering-related
|
Research Institution | The University of Tokyo |
Principal Investigator |
Mitsuya Yuki 東京大学, 大学院工学系研究科(工学部), 助教 (70784825)
|
Project Period (FY) |
2019-04-01 – 2022-03-31
|
Keywords | CT / 圧縮センシング / 非破壊検査 / 最適化数学 |
Outline of Final Research Achievements |
In this study, a method for reconstructing CT results projected from a translational trajectory was investigated. Linear orbit projection uses the spread angle of the X-ray source to obtain a parallax image. However, in this case, the angular information is insufficient compared to the conventional rotational orbit projection. Therefore, the conventional reconstruction algorithm cannot provide a perfect reconstruction result. In this study, we demonstrated translational CT using the reconstruction algorithm of compressed sensing-based CT. In addition to total variation regularization, we used a new concept of directional difference regularization method, and constructed a reconstruction algorithm. An experimental system was actually constructed and CT demonstration was successfully performed.
|
Free Research Field |
放射線計測
|
Academic Significance and Societal Importance of the Research Achievements |
回転軌道での撮像ではなく、直線的な軌道での投影からCTを行うことができれば、大型構造物に対しても断層撮像を行うことができるため、望ましい。この直線軌道CTでは、コーンビームやファンビームのX線が発散する角度をうまく利用して、視差のあるX線画像を取得して、そこから断面画像の再構成を行うというコンセプトである。直線軌道CTは広範囲な大型構造物、移動が困難な物体に対しても適用できると考えられるため、新しいCT技術として産業や公益面で大きなインパクトがあると考えられる。
|