A Deformable Weight Content-Based Image Retrieval Using Color Features
碩士 === 國立中山大學 === 資訊工程研究所 === 87 === Traditionally, content based image retrieval focuses on the complete image matching and retrieval. It is difficult to match and retrieve part of the image. In this thesis, we present a deformable weight color matching and a color feature extraction met...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | en_US |
| Published: |
1999
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/52124620190622519532 |
| Summary: | 碩士 === 國立中山大學 === 資訊工程研究所 === 87 === Traditionally, content based image retrieval focuses on the complete image matching and retrieval. It is difficult to match and retrieve part of the image. In this thesis, we present a deformable weight color matching and a color feature extraction method to solve the problem. There are two kinds of the similarity measurement methods used in the prototype system. One is the complete image similarity measurement, and the other is the deformable weight image similarity measurement. We also use a hierarchical filtering at different levels to reduce the computational complexity. We use modified pyramid scheme to decompose source image into four sub images in the next level for the progressive transform image. The client first transmits a part of the source image. Until the user asks for higher quality image, the rest of image are transmitted to the client and composed to a high quality image.
|
|---|