| Project/Area Number |
17500127
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Perception information processing/Intelligent robotics
|
|---|
| Research Institution | Ryukoku University |
|---|
Principal Investigator |
FUJITA Kazuhiro Ryukoku University, Faculty of Science and Technology, Associate Professor, 理工学部, 助教授 (90209049)
|
|---|
| Project Period (FY) |
2005 – 2006
|
| Project Status |
Completed (Fiscal Year 2006)
|
| Budget Amount *help |
¥3,000,000 (Direct Cost: ¥3,000,000)
Fiscal Year 2006: ¥1,300,000 (Direct Cost: ¥1,300,000)
Fiscal Year 2005: ¥1,700,000 (Direct Cost: ¥1,700,000)
|
|---|
| Keywords | Camera Shake / Motion-Blur / JPEG / Parameter Estimation / Image Restoration |
|---|
| Research Abstract |
A method of improving motion-blurred JPEG image is proposed. The model about JPEG coding / decoding process, edge orientation and local activity are taken into account to conventional regularized image restoration method. Edge orientation is extracted from temporally improved image by conventional regularized image restoration method in which the JPEG coding / decoding process is not taken into account. The result of computer simulation shows effectiveness of this proposed method.
Before improving process, the parameter of motion-blur is necessary. But estimating the parameter of motion-blur is difficult Therefore instead of estimating the parameter of motion-blur, evaluating the improved motion-blurred JPEG image is proposed. The evaluation value of the improved motion-blurred JPEG image is the skewness of the spectrum for the output image by applying measurement differentiating filter to the improved motion-blurred JPEG image. The above evaluation value of the improved motion-blurred JPEG image is based on both the idea that the measurement differentiating filter acts like the whitening filter, and the idea that suitably improved image is made to be white by the measurement differentiating filter. The result of computer simulation shows that the proposed evaluation value of the improved motion-blurred JPEG image acts like mean square error between the original image and improved motion-blurred JPEG image.
|
