Abstract | ||
---|---|---|
Because there is not a perfect feature descriptor which can suit to describe any image, this paper proposes combining different feature descriptor to do the content-based image retrieval and using the LIRe to implement a image retrieval system. The experiment shows that using multi-feature descriptor at the same time can keep the retrieval system having a stable accuracy. |
Year | DOI | Venue |
---|---|---|
2012 | 10.1109/ICDMA.2012.237 | ICDMA |
Keywords | Field | DocType |
retrieval system,implement image,multi-feature descriptor,perfect feature descriptor,stable accuracy,content-based image retrieval,image retrieval system,different feature descriptor,indexes,image retrieval,histograms,feature extraction,accuracy | Histogram,Computer vision,Automatic image annotation,Feature detection (computer vision),Pattern recognition,GLOH,Computer science,Image texture,Image retrieval,Feature extraction,Artificial intelligence,Visual Word | Conference |
Citations | PageRank | References |
2 | 0.39 | 6 |
Authors | ||
2 |