2014 Fiscal Year Final Research Report
Joint image transmission via convex optimization
| Project/Area Number |
24560443
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Communication/Network engineering
|
|---|
| Research Institution | Chiba Institute of Technology |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2012-04-01 – 2015-03-31
|
| Keywords | 凸最適化 / 画像符号化 / 画像処理 |
|---|
| Outline of Final Research Achievements |
Convex optimization based methods are widely proposed in image processing research field. Such methods utilize prior information of desired result by representing the information as a convex objective function. In this research project, we tried to push the boundary of the convex optimization based methods towards three important problems --- image inpainting, image retargeting and image coding--- which are the building blocks of end-to-end image transmission system. We proposed a low-rank matrix based image inpainting method, an image retargeting method by focusing on ROI (region of interest) movement, and a post processing method for coded chrominance component by using the luminance-chrominance correlation.
|
| Free Research Field |
画像処理
|
