Abstract | ||
---|---|---|
In this paper, we propose a term selection model to help select terms in the documents that describe the images to improve the content-based image retrieval performance. First, we introduce a general feature selection model. Second, we present a painless way for training document collections, followed by selecting and ranking the terms using the Kullback-Leibler Divergence. After that, we learn the terms by the classification method, and test it on the content-based image retrieval result. Finally, we setup a series of experiments to confirm that the model is promising. Furthermore, we suggest the optimal values for the number maxK and the tuning combination parameter α in the experiments. |
Year | DOI | Venue |
---|---|---|
2010 | 10.1007/978-3-642-15470-6_40 | AMT |
Keywords | Field | DocType |
optimal value,term selection model,training document collection,classification method,number maxk,kullback-leibler divergence,select term,content-based image retrieval performance,general feature selection model,content-based image retrieval result,machine learning,feature selection,kullback leibler divergence | Divergence-from-randomness model,Data mining,Feature selection,Computer science,Image retrieval,Artificial intelligence,Automatic image annotation,Information retrieval,Pattern recognition,Ranking,Machine learning,Content-based image retrieval,Visual Word | Conference |
Volume | ISSN | ISBN |
6335.0 | 0302-9743 | 3-642-15469-7 |
Citations | PageRank | References |
1 | 0.35 | 18 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Qinmin Vivian Hu | 1 | 20 | 6.06 |
Zheng Ye | 2 | 45 | 3.01 |
Xiangji Huang | 3 | 1551 | 159.34 |