2009 Fiscal Year Final Research Report
Sampling theory for sparse signals and its applications to image super-resolution
Project/Area Number |
20700164
|
Research Category |
Grant-in-Aid for Young Scientists (B)
|
Allocation Type | Single-year Grants |
Research Field |
Perception information processing/Intelligent robotics
|
Research Institution | Yamaguchi University |
Principal Investigator |
HIRABAYASHI Akira Yamaguchi University, 大学院・医学系研究科, 准教授 (50272688)
|
Co-Investigator(Renkei-kenkyūsha) |
PIER-LUIGI Dragotti ロンドンインペリアル大学
|
Project Period (FY) |
2008 – 2009
|
Keywords | スパース性 / 標本化 / マルチチャンネル標本化 / MAP推定 / l1ノルム最小化 / 単位長自由度有限信号 / 画像特徴量抽出 |
Research Abstract |
We studied multi-channel sampling for sparse signals and image feature extraction using the sampling scheme. The former was mainly tackled in the first year, and we proposed two algorithms ; one is the so-called MAP estimation algorithm with a prior distribution of signals, and the other is the l1-norm minimization algorithm. In the second year, we devised an exact line-edge extraction algorithm that can be applied for super-resolution. We showed by computer simulations that the proposed method is approximately 10 dB better against noise than the conventional methods.
|