A Study on Image Classification using Hybrid Method

Park Sangsung; JANG， Dong Sik; 정귀임

Journal of The Korea Society of Computer and Information

Abbr : JKSCI
2006, 11(6), pp.79-86
Publisher : The Korean Society Of Computer And Information
Research Area : Computer Science

Park Sangsung ¹, JANG， Dong Sik ², 정귀임 ³

¹고려대학교
²고려대학교
³고려대학교

ABSTRACT

Classification technology is essential for fast retrieval in large multi-media database. This paper proposes a combining GA(Genetic Algorithm) and SVM(Support Vector Machine) model to fast retrieval. We used color and texture as feature vectors. We improved the retrieval accuracy by using proposed model which retrieves an optimal feature vector set in extracted feature vector sets. The first performance test was executed for the performance of color, texture and the feature vector combined with color and texture. The second performance test, was executed for performance of SVM and proposed algorithm. The results of the experiment, using the feature vector combined color and texture showed a good performance than a single feature vector and the proposed algorithm using hybrid method also showed a good performance than SVM algorithm.
KEYWORDS

Hybrid method, Data mining, GA(Genetic Algorithm), SVM(Support Vector Machine), Image, Classification
REFERENCES (8)

[journal] / 1996 / Tools and techniques for color image retrieval 2670 : 426~437

[journal] / 1995 / Texture features for browsing and retrieval of image data

[journal] / 1998 / Shape-based retrieval: A case study with trademark image databases 31(9) : 1369~1390

[journal] / 1995 / Query by image content: The QBIC system 28(9) : 23~31

[journal] / 1996 / VisualSEEK: A fully automated content-based image query system : 87~98

[journal] / 2006 / “유전자 알고리즘을 이용한 영상 특징 추출” 11(3) : 133~139

[journal] / 1998 / Statistical Learning Theory / New York

[journal] / 1994 / Support vector machines for spam catergori-zation 10(5) : 1048~1054
KCI Citation (0)

KJC Korea
Journal Central

Statistics

Tools