.png){kind=icon}
.png){kind=icon}
Return
Improving distance based image retrieval using non-dominated sorting genetic algorithm
.png){kind=icon}
PRE
AI
DOI：10.1016/j.patrec.2014.05.008.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Relevance feedback has been adopted as a standard in Content Based Image Retrieval (CBIR). One major difficulty that algorithms have to face is to achieve and adequate balance between the exploitation of already known areas of interest and the exploration of the feature space to find other relevant areas. In this paper, we evaluate different ways to combine two existing relevance feedback methods that place unequal emphasis on exploration and exploitation, in the context of distance based methods. The hybrid approach proposed has been evaluated by using three image databases of various sizes that use different descriptors. Results show that the hybrid technique performs better than any of the original methods, highlighting the benefits of combining exploitation and exploration in relevance feedback tasks. (C) 2015 Elsevier B.V. All rights reserved.
Keywords:
Image retrieval
Multiobjective genetic algorithm
Relevance feedback
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
3.3
Papers:
7.9K
Citations:
1.6W