A PARAMETER RE-RANKING TECHNIQUE FOR WEB IMAGE SEARCH
| Author(s) | : | veladandi divya |
| Institution | : | Computer Science And Engineering, Svs Engineering College |
| Published In | : | Vol. 4, Issue 11 — November 2017 |
| Page No. | : | 260-262 |
| Domain | : | Engineering |
| Type | : | Research Paper |
| ISSN (Online) | : | 2348-4470 |
| ISSN (Print) | : | 2348-6406 |
We recommend making use of strong semantic relationship within graph for the image search re-rankingand exploit semantic attributes in support of image search re-ranking and on the basis of classifiers for the entirepredefined attributes, each of the image is symbolised by an attribute feature that includes responses from theseclassifiers. Semantic attributes have received more consideration in the recent times in which efficiency was verified inbroad applications. Semantic attributes is considered as mid-level semantic preserving notion. We introduce a hypergraph to model relationship among images by means of integration of low-level features as well as attribute features andlater hyper-graph ranking is carried out to order the images and its fundamental principle is that visually similar imagesinclude related scores of ranking. We introduce attribute-assisted re-ranking technique that is based on hyper-graphlearning and we instruct numerous classifiers for the entire pre-defined attributes and every image is represented bymeans of attribute feature that includes responses from these classifiers.
veladandi divya, “A PARAMETER RE-RANKING TECHNIQUE FOR WEB IMAGE SEARCH”, International Journal of Advance Engineering and Research Development (IJAERD), Vol. 4, Issue 11, pp. 260-262, November 2017.
