Image classification it generally requires a priori knowledge about the objects to be classified. In this paper, we present a new method to segment tumor in multispectral magnetic resonance (MR) images of the human brain. The proposed approach, called Minimum Variance Distortionless Response beamforming (MVDR) was introduced in [15] where only the knowledge of the desired signature to be classi...

This paper proposes a stereo acoustic echo suppression method for duplex stereo teleconferencing. Firstly the stereo signals are combined in the frequency domain to build a single-channel signal. Secondly, with the single-channel signal model, an optimal Linearly Constrained Minimum Variance (LCMV) filter based on the Widely Linear theory is derived, which is applied on microphone signal direct...

This paper introduces the constrained version of the Affine Projection Algorithm. The new algorithm is suitable for linearly-constrained minimum-variance applications, which include beamforming and multiuser detection for communications systems. The paper also discusses important aspects of convergence and stability of constrained normalized adaptation algorithms in general. It is shown that no...

The minimum average correlation energy filter (MACE) [1] [5] is of interest to the ATD/R problem due to its inherent properties. As an image from the recognition class becomes centered on the filter mask, the MACE filter produces a sharp peak and as the image moves away from the center of the filter a low variance output results. The filter can be modified to produce a low variance output for a...

A pattern synthesis method for arbitrary arrays based on the linearly constrained minimum variance (LCMV) criterion is presented. Given mainlobe regions and an arbitrary sidelobe envelope, this algorithm searches the pattern with the lowest sidelobe levels. Its iteration coefficient is robust to synthesis conditions, and patterns with a flat top mainlobe can be obtained using phase-independent ...

This paper proposes a new approach to linearly-constrained adaptive filtering, where successive Householder transformations are incorporated in the algorithm update equation in order to reduce computational complexity and coefficienterror norm. We show the derivation of two new algorithms, namely the unnormalized and the normalized Householdertransform constrained LMS algorithms (HCLMS and NHCL...

Independent Component Analysis (ICA) designed for complete bases is used in a variety of applications with great success, despite the often questionable assumption of having N sensors and M sources with NQM. In this article, we assume a source model with more sources than sensors (M>N), only L<N of which are assumed to have a non-Gaussian distribution. We argue that this is a realistic source m...

This paper aims at presenting our algorithm used to make submission for the NIST 2013-2014 speaker recognition ivector challenge. The fixed dimensional i-vector representation of speech utterances has attracted attentions from other communities. This challenge focuses on the task of speaker detection using i-vectors derived from conversational telephony speech data. However, the unlabeled i-vec...