2021 Fiscal Year Annual Research Report
劣化したハイパースペクトル光音響イメージの復元に関する研究
| Project/Area Number |
21J13438
|
|---|
| Research Institution | Tokyo Institute of Technology |
|---|
Principal Investigator |
LIAO WEIHANG 東京工業大学, 工学院, 特別研究員(DC2)
|
|---|
| Project Period (FY) |
2021-04-28 – 2023-03-31
|
| Keywords | Photo-acoustic image / Hyper-spectral PA / Restoration |
|---|
| Outline of Annual Research Achievements |
During the past year, we have achieved the following research progress: First, we proposed a HSPA reliability analysis. We captured PA data of carefully designed 3D printed objects, and investigated the special feature that observations on different parts have dissimilar reliabilities depending on the absorbed light energy; Second, we introduced graph model for PA data representation, in which we explore the spatial, spectral, and depth correlation simultaneously in 4D space; Third, we formulated the restoration task into an optimization problem. We explore global sparsity and local self-similarity as data constraints and solve the optimization via numerical algorithm; Last, we finished the prototype of the 4D HSPA restoration framework, and published it on a top international conference.
|
| Current Status of Research Progress |
Current Status of Research Progress
2: Research has progressed on the whole more than it was originally planned.
Reason
Currently, under the supervision of my academic advisor, this research project is progressing smoothly as expected. We have finished the 1st year research plan as we wrote in the application form.
We have one top international conference paper (IEEE/CVF Conference on Computer Vision and Pattern Recognition) accepted, one journal paper and another conference paper are submitted, now are under revision.
|
| Strategy for Future Research Activity |
In the next year, we will follow the original research plan as we wrote in the application form. We plan to:
First, consider other type of degradation; Second, capture more HSPA data; Third, improve the whole framework performance; Fourth, consider the extension to 5D HSPA video.
|
