| Presentation | 2014-07-10 Image Recovery by Using Median Regularization Makoto NAKASHIZUKA, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In this paper, a regularization method based on local median is proposed for image recovery problem. The image prior of the regularization is defined from the sufficient condition that a pixel is a median among its neighboring pixels. The objective function of the regularization with the proposed median prior is approximated as a form of a weighted least square problem. IRLS (Iteratively reweighted least square) algorithm is applied for image recovery with the proposed regularization. Each iteration corresponds to the Tikhonov regularization with the Laplacian, of which coefficients are adapted along with the median prior. In image recovery experiments, image denoising and single image super-resolution are demonstrated. Comparisons of the proposed prior with total variation (TV) prior are also shown. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Median filter / image recovery / regularization / super-resolution / iteratively reweighted least square |
| Paper # | CAS2014-25,VLD2014-34,SIP2014-46,MSS2014-25,SIS2014-25 |
| Date of Issue |
| Conference Information | |
| Committee | MSS |
|---|---|
| Conference Date | 2014/7/2(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Mathematical Systems Science and its applications(MSS) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Image Recovery by Using Median Regularization |
| Sub Title (in English) | |
| Keyword(1) | Median filter |
| Keyword(2) | image recovery |
| Keyword(3) | regularization |
| Keyword(4) | super-resolution |
| Keyword(5) | iteratively reweighted least square |
| 1st Author's Name | Makoto NAKASHIZUKA |
| 1st Author's Affiliation | Faculty of Engineering, Chiba Institute of Technology() |
| Date | 2014-07-10 |
| Paper # | CAS2014-25,VLD2014-34,SIP2014-46,MSS2014-25,SIS2014-25 |
| Volume (vol) | vol.114 |
| Number (no) | 125 |
| Page | pp.pp.- |
| #Pages | 5 |
| Date of Issue | |