| Project/Area Number |
20360178
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
System engineering
|
|---|
| Research Institution | Osaka Prefecture University |
|---|
Principal Investigator |
OMATU Sigeru Osaka Prefecture University, 工学部, 教授 (30035662)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
FUJINAKA Toru 広島大学, 大学院・教育学研究科, 教授 (90190058)
YOSHIOKA Michifumi 大阪府立大学, 大学院・工学研究科, 教授 (70285302)
FUJIMURA Masao 大阪工業大学, 工学部, 准教授 (80319574)
|
|---|
| Project Period (FY) |
2008 – 2010
|
| Project Status |
Completed (Fiscal Year 2010)
|
| Budget Amount *help |
¥18,590,000 (Direct Cost: ¥14,300,000、Indirect Cost: ¥4,290,000)
Fiscal Year 2010: ¥5,330,000 (Direct Cost: ¥4,100,000、Indirect Cost: ¥1,230,000)
Fiscal Year 2009: ¥6,240,000 (Direct Cost: ¥4,800,000、Indirect Cost: ¥1,440,000)
Fiscal Year 2008: ¥7,020,000 (Direct Cost: ¥5,400,000、Indirect Cost: ¥1,620,000)
|
|---|
| Keywords | 音響・画像の融合化 / 特徴抽出 / ニューラルネットワーク / 紙幣識別 / 紙幣の新旧識別 / 高精細画像生成 / 超解像 / 音響・画像統合化 / 3次元画像 / ハルシネーション / イメージハルシネーション / 低解像度画像 / 高解像度画像 / 局所特徴量 / キャニーフィルタ / ダウンサンプリング / アップサンプリング |
|---|
| Research Abstract |
This study is to develop an intelligent acoustic diagnostics systems using intelligent systems technology and extend their application fields to not only acoustic inspection but also visual inspection using images measured by industrial instruments. The originality of the research is to marge two types of information such as acoustic data and image data at the same time and to process them as a multi-dimensional data. In order to develop the intelligent systems, we have used the evolutional computational techniques such as neural networks, fuzzy technology, genetic algorithms, genetic programming, soft-computing, etc.
The intelligent technology developed here has been applied to classification of bills using a banknote counter machine. The performance to the classification is to achieve 100 % with misclassification rate of 10^<-12> and the specification including both kinds of bills and truth or falsehood of bills. Furthermore, the processing speed is 30 pieces per second. Those efficiencies have been achieved by the proposed approach.
The other applications such as the inspection of securities at stock markets, check at banks, and hand written letter recognition has been adopted to show the effectiveness in a real market of the proposed technology.
|
