IPIU 1997 (제9회 영상처리 및 이해에 관한 워크샵)
Current Result Document :
| 한글제목(Korean Title) |
복합 그래프 표현을 이용한 컬러 영상의 검색 기법 |
| 영문제목(English Title) |
Color Image Retrieval Using Hybrid Graph Representation |
| 저자(Author) |
박인규
윤일동
이상욱
In Kyu Park
Il Dong Yun
Sang Uk Lee
|
| 원문수록처(Citation) |
VOL 09 NO. 01 PP. 0197 ~ 0202 (1997. 01) |
한글내용
(Korean Abstract) |
|
영문내용
(English Abstract) |
In this paper, an effective color image retrieval algorithm is proposed based on a hybrid graph representation, which consists of Modified Color Adjacency Graph (MCAG) and Spatial Variance Graph (SVG). The MCAG, which improves Color Adjacency Graph (CAG), is proposed to enhance the matching sensitivity and managing ability for a large image database, in which color adjacency information as well as color histogram is taken into account. In addition, the SVG is introduced to consider the geometric information in spatial domain. In the SVG, self and mutual variances are adopted as the attributes of each pair of effective nodes of MCAG so that spatial relationship between distinct nodes is properly modeled. After generating graphs, graph matching is performed to retrieve the best matched image by using the proposed similarity measure. Experiments are performed for Swain’s test images and VIRAGE images, which show that the proposed algorithm provides high retrieval performance, compared to the previous works.
|
| 키워드(Keyword) |
|
| 파일첨부 |
PDF 다운로드
|
