Most methods for the detection and removal of specular reflections suffer from nonuniform highlight regions and/or nonconverged artifacts induced by discontinuities in the surface colors, especially when dealing with highly textured, multicolored images. In this paper, a novel noniterative and predefined constraint-free method based on tensor voting is proposed to detect and remove the highligh...

The K Nearest Neighbors classification method assigns to an unclassified observation the class which obtains the best results after a voting criteria is applied among the observation’s K nearest, previously classified points. In a validation process the optimal K is selected for each database and all the cases are classified with this K value. However the optimal K for the database does not hav...

We consider two classification approaches. The metric-based approach induces the distance measure between objects and classifies new objects on the basis of their nearest neighbors in the training set. The rule-based approach extracts rules from the training set and uses them to classify new objects. In the paper we present a model that combines both approaches. In the combined model the notion...

The current development of e-voting systems worldwide raises several specific interesting issues from a legal point of view. Auditability measures, identification procedures or guarantees for voting secrecy and equality are good examples, but we often forget a fundamental question: the usefulness of these new technologies. This paper intends to provide an answer that takes into account the comp...

It is debatable if current direct-recording electronic voting machines can sufficiently be trusted for a use in elections. Reports about malfunctions and possible ways of manipulation abound. Voting schemes have to fulfill seemingly contradictory requirements: On one hand the election process should be verifiable to prevent electoral fraud and on the other hand each vote should be deniable to a...

With the rapid increase of protein sequences in the post-genomic age, it is challenging to develop accurate and automated methods for reliably and quickly predicting their subcellular localizations. Till now, many efforts have been tried, but most of which used only a single algorithm. In this paper, we proposed an ensemble classifier of KNN (k-nearest neighbor) and SVM (support vector machine)...

Several eeective methods for improving the performance of a single learning algorithm have been developed recently. The general approach is to to create a set of learned models by repeatedly applying the algorithm to diierent versions of the training data, and then combine the learned models' predictions according to a prescribed voting scheme. Little work has been done in combining the predict...

The goal of traditional multi-instance learning (MIL) is to predict the labels of the bags, whereas in many real applications, it is desirable to get the instance labels, especially the labels of key instances that trigger the bag labels, in addition to getting bag labels. Such a problem has been largely unexplored before. In this paper, we formulate the Key Instance Detection (KID) problem, an...

Solar images taken at different wavelengths enable scientists to visualize and analyze the suns activities. The Solar Dynamics Observatory (SDO) provides high-resolution images of the sun, with cadence in seconds, taken at varying wavelengths, resulting in finely detailed, almost continuous data for researcher’s examination. We propose an approach to find active regions and coronal holes that i...

Classification problems with an imbalanced class distribution have received an increased amount of attention within the machine learning community over the last decade. They are encountered in a growing number of real-world situations and pose a challenge to standard machine learning techniques. We propose a new hybrid method specifically tailored to handle class imbalance, called EPRENNID. It ...