Spike Timing Dependent Plasticity (STDP) is a learning rule that modifies synaptic strength as a function of the relative timing of pre-and postsynaptic spikes. Here we use this learning rule with neurons integrating spike trains coming from V1 orientation selective cells. Presenting natural images containing faces we observe that the neurons develop selectivity to face features. These results ...

In this paper, we propose a novel cascading approach, by cascading the feature selection method using mutual correlation with this symbolic approach. In the symbolic approach, the new dimensionality reduction method through transformation of features into symbolic data using the property of collinearity and variance based cumulative sum of features is used. The feature values are transformed in...

This paper presents a new approach to noise reduction and contrast enhancement for different types of medical images. An anisotropic scheme is used to iteratively reduce noise as well as to define image regions and enhance region contrast. The process is performed in three stages and the final values of the regions are automatically extracted from the image histogram, thus providing a fast meth...

On the basis of manifold learning theory, a new feature extraction method for Synthetic aperture radar (SAR) target recognition is proposed. First, the proposed algorithm estimates the within-class and between-class local neighbourhood surrounding each SAR sample. After computing the local tangent space for each neighbourhood, the proposed algorithm seeks for the optimal projecting matrix by pr...

We present a framework for enhancing images while pre serving either the edge or the orientation dependent texture informa tion present in them We do this by treating images as manifolds in a feature space This geometrical interpretation leads to a natural way for grey level color movies volumetric medical data and color texture im age enhancement Following this we invoke the Polyakov action fr...

We present a fast video retrieval system with three novel characteristics. First, it exploits the methods of machine learning to construct automatically a hierarchy of small subsets of features that are progressively more useful for indexing. These subsets are induced by a new heuristic method called SortMerge feature selection, which exploits a novel combination of Fastmap for dimensionality r...

In this paper, a fusion of two methods for dimensionality reduction is proposed. First method is the selection of features using FQ measure method followed by another method based on symbolic approach is proposed. The symbolic method is based on the transformation of features into symbolic data using the property of collinearity and variance based cumulative sum of features. In this proposed ap...

Transfer learning addresses the problem of how to utilize plenty of labeled data in a source domain to solve related but different problems in a target domain, even when the training and testing problems have different distributions or features. In this paper, we consider transfer learning via dimensionality reduction. To solve this problem, we learn a low-dimensional latent feature space where...

Image feature selection is an important issue for source camera identification. Well-selected features should make camera classifiers accurate, efficient as well as robust. Current source camera identification schemes select image features mainly based on classification accuracy and computational efficiency. In this work, we demonstrate that robustness should also be considered for classifiers ...

The rapid advance of computer technologies in data processing, collection, and storage has provided unparalleled opportunities to expand capabilities in production, services, communications, and research. However, immense quantities of high-dimensional data renew the challenges to the state-of-the-art data mining techniques. Feature selection is an effective technique for dimension reduction an...