A novel Bayesian framework for relevance feedback in image content-based retrieval systems

Applying logistic regression to relevance feedback in image retrieval systems

“…By solving a minimization problem on the parameter estimation process, the authors conclude that the best M matrix (restricted to diagonal matrices only) is given by m ij ∝ 1/ 2 j , j being the standard deviation of the jth vector component. Unlike the present paper, our previous work concerning relevance feedback CBIR algorithms was focused on a Bayesian strategy [15]. We followed the idea of modeling user preferences as a probability distribution.…”

A relevance feedback CBIR algorithm based on fuzzy sets

“…The ceramic tile category has been introduced in order to offer a large diversity of feature vectors, a fact which hinders the search process by introducing nondesirable elements into the results obtained at each iteration. This same repository was used in the evaluation of [28], and a subset was employed in [14].…”

“…Because of the small number of images in the repository, we have used a standard SOM instead of the tree-structured SOM proposed in the original publication. (3) The Bayesian technique described in [28]. This is a recent algorithm developed by our research group, belonging to the probabilistic-based approaches.…”

“…Both memory requirements and execution time scale linearly with the number of images in the database. Another similar approach was presented in [28], based on the modeling of user preferences as a probability distribution on the image space. This distribution is the prior distribution and its parameters are modified based on the information provided by the user.…”

Content Based Image Retrieval using Novel Gaussian Fuzzy Feed Forward-Neural Network

“…Classifiers based on Bayesian network (Cox et al, 2000;De Ves et al, 2006), Self organizing maps (Koskela et al, 2004), Support vector machine (Zhou and Huang, 2001;Tong and Chang, 2001) and Regression models (Leon et al, 2007) have been proposed with varying classification accuracy. Ghrare et al (2009) proposed lossless coding for the image retrieval problem by using lossless compression with high accuracy.…”