A New Approach to Interactive Visual Search with RBF Networks based on Preference Modelling

In this paper, we propose a new method for image retrieval with relevance feedback based on eliciting preferences from the decision maker aquiring visual information from an image database. The proposed extension of the common approach to image retrieval with relevance feedback allows it to be applied to objects with non-homogenous colour and texture. This has been accomplished by the algorithms, which model user queries by an RBF neural network. As an example of the application of this approach, we have used a content-based search in an atlas of species. An experimental comparision with the commonly-used content-based image retrieval approach is presented.

ROTTER Pawel; 

2009-01-23

SPRINGER-VERLAG BERLIN

JRC49944

0302-9743,   

http://www.springerlink.com/content/y6r4442071062h47/?p=c66e3a5f3b5a410796050413162d700d&pi=81,    https://publications.jrc.ec.europa.eu/repository/handle/JRC49944,   

10.1007/978-3-540-69731-2,