| Project/Area Number |
16K16091
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Perceptual information processing
|
|---|
| Research Institution | Tokai University (2018)
Tokyo University of Science (2016-2017) |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2016-04-01 – 2019-03-31
|
| Project Status |
Completed (Fiscal Year 2018)
|
| Budget Amount *help |
¥3,770,000 (Direct Cost: ¥2,900,000、Indirect Cost: ¥870,000)
Fiscal Year 2018: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2017: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2016: ¥2,470,000 (Direct Cost: ¥1,900,000、Indirect Cost: ¥570,000)
|
|---|
| Keywords | 画像修復 / 信号復元 / 一般化主成分分析 / ランク最小化 / スパース最適化 / スパースモデリング / スイッチングモデル / 同時最適化 / 信号修復 / 部分空間クラスタリング / TV最小化 / 情報通信工学 / 感性情報学 / 画像 / モデル化 |
|---|
| Outline of Final Research Achievements |
In this study, an image repair problem in which an arbitrary part of an image is regarded as a missing part and the missing part is restored from the information of the circumference was handled, and it aimed at the construction of an image repair technique with high accuracy based on the generalized principal component analysis.
Research carried out during the study period and its results are (1) a proposal of an image repair algorithm that is robust to model deviation using sparse optimization, and (2) a proposal of a signal repair method using generalized principal component analysis.
One peer-reviewed paper was adopted for (1), and one peer-reviewed paper and 2 conferences awards for (2) were highly evaluated.
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究のようにスイッチングモデルを仮定し，最適化によってモデルの推定，ピクセルとモデルの対応関係推定，欠損ピクセルの修復の３つを同時に行う手法は他に無く，この点に特色と独創性がある．本研究によって画像修復精度が向上し，商用写真加工の現場でのコスト軽減が期待される．また，本研究で確立されるいくつかの手法は，画像修復のみにとどまらず，信号修復理論一般へ波及すると考えられ，その点からも本テーマの意義は大きい．
|
