2017 Fiscal Year Final Research Report
Real-time personal identification at night
| Project/Area Number |
26330190
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Perceptual information processing
|
|---|
| Research Institution | Niigata University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2014-04-01 – 2018-03-31
|
| Keywords | パターン認識 / 個人認証 / 夜間 / Kinect / 歩行動作 / 身体サイズ |
|---|
| Outline of Final Research Achievements |
This research proposed a personal identification method using body motion data and size data obtained from the Kinect sensor. This method is used in the night. From the motion data, personal identification is performed in an eigenspace smaller than the actual parameter space. In this case, the calculation efficiency is further improved by using several 2D eigenplanes. This method is called a higher order eigenspace method. On the other hand, from the size data, the nearest neighbor method based on Euclidean distance has the highest identification rate. In addition, the identification rate is more improved by a strong classifier to be determined by the minimum sum of ranks from the higher order eigenspace method and the nearest neighbor method.
|
| Free Research Field |
知覚情報処理
|
