Browse > Article
http://dx.doi.org/10.9708/jksci.2010.15.11.075

Edge Feature Extract CBIRS for Car Retrieval : CBIRS/EFI  

Koo, Gun-Seo (숭의여자대학 인터넷정보과)
Publication Information
Journal of the Korea Society of Computer and Information / v.15, no.11, 2010 , pp. 75-82 More about this Journal
Abstract
The paper proposed CBIRS/EFI with contents based search technique using edge feature information of the object from image information of the object which is uncertain. In order to search specially efficiently case of partial image information of the object, we used the search technique which extracts outline information and color information in feature information of object. In order to experiment this, we extracted side edge feature information of the vehicle for feature information of the object after capture the car image of the underground garage. This is the system which applies a contents base search by the result which analyzes the image which extracts a feature, an original image to search and a last similar measurement result. This system compared in FE-CBIRS systems which are an existing feature extraction contents base image retrieval system and the function which improves the accuracy and an effectiveness of search rate was complemented. The performance appraisal of CBIRS/EFI systems applied edge extraction feature information and color information of the cars. And we compared a color feature search time, a shape characteristic search time and a search rate from the process which searches area feature information. We extracted the case 91.84% of car edge feature extraction rate. And a average search time of CBIRS/EFI is showing a difference of average 0.4-0.9 seconds than FE-CBIRS from vehicle. color search time, shape characteristic search time and similar search time. So, it was proven with the fact that is excellent.
Keywords
FE-CBIRS:(Feature Extraction-Content Based Image Retrieval System); H.S.I Color Space; CBIRS/EFI(Content Based Image Retrieval System for Edge Feature Information);
Citations & Related Records
Times Cited By KSCI : 3  (Citation Analysis)
연도 인용수 순위
1 최형일, "색상과 형태를 이용한 내용기반 영상 검색," 한국컴퓨터정보학회 제 13권, 제 1호, 117-124쪽, 2008년 1월.   과학기술학회마을
2 백두원, "관심영역을 고려한 색 양자화 방법," 한국컴퓨터정보학회 제 12권, 제 6호, 161-165쪽, 2007년 12월.   과학기술학회마을
3 김봉기, "멀티미디어 데이터베이스를 위한 2단계 내용기반 영상 검색 기법," 숭실대학교 박사학위 논문, 1998년.
4 서영건, 박순화, 이부건, "피사계 심도가 낮은 이미지에서 위비블릿 기반의 자동 관심 영역 추출", 한국 컴퓨터정보학회 동계학술대회 논문집 제 17권, 제 1호, 2008년 1월.
5 구건서, "IPTV에서 컷 검색을 위한 색 분포 정보를 이용한 FE-CBIRS," 한국컴퓨터정보학회 제 14권, 제 1호, 91-97쪽, 2009년 1월.   과학기술학회마을
6 이동호, 유광석, 김회율, "컬러와 모양 정보를 이용한 캐 릭터 이미지 검색," 방송공학회논문지, 제 5권, 제 1호, 50-60쪽, 2000년 6월.
7 Bojkovic, Zoran; Samcovic, Andreja; "Face Detection Approach in Neural Network Based Method for Video Surveillance," Neural Network Applications in Electrical Engineering, NEUREL 2006. pp. 44-47, Sept. 2006
8 Charles Frankel, Michael J. Swai, and Vassilis Athitsos, "WebSeer: An Image Search Engine for the World Wide Web," Technical Report 96-14, 2006.
9 Chih-ChangChen; "Automatically Determined Region of Interest in JPEG 2000," Multimedia, IEEE Trans, Vol. 9-7, pp.1333-1345, Nov. 2007.