Search results for "Pattern Recognition"
showing 10 items of 2301 documents
Spectral interest points and texture extraction and fusion for identification, control and security
2018
Biometrics is an emerging technology that proposes new methods of control, identification and security. Biometric systems are often subject to risks. Face recognition is popular and several existing approaches use images in the visible spectrum. These traditional systems operating in the visible spectrum suffer from several limitations due to changes in lighting, poses and facial expressions. The methodology presented in this thesis is based on multispectral facial recognition using infrared and visible imaging, to improve the performance of facial recognition and to overcome the deficiencies of the visible spectrum. The multispectral images used in this study are obtained by fusion of visi…
Non-linear RLS-based algorithm for pattern classification
2006
A new non-linear recursive least squares (RLS) algorithm is presented in the context of pattern classification problems. The algorithm incorporates the non-linearity of the filter's output in the updating rules of the classical RLS algorithm. The proposed method yields lower stationary error levels when compared to the standard LMS and RLS algorithms in a classical application of pattern classification, such as the channel equalization problem.
Tetrolet-based reduced reference image quality assessment approach
2014
In this paper, we propose a new reduced reference image quality assessment (RRIQA) scheme. For this purpose, we use a statistical-based method in a new adaptive Haar wavelet transform domain, called Tetrolet. Firstly, we decompose the reference and distorted images and we obtain the Tetrolet coefficients for each image. Secondly, we use a marginal Generalized Gaussian Density (GGD) to model each subband coefficients. Finally, the distortion measure is computed using the Kullback Leibler Divergence (KLD) between GGD Probability density function (PDFs). Experimental results show the efficiency of the proposed method when comparing to those reported in the literature.
3D Reconstruction of Dynamic Vehicles using Sparse 3D-Laser-Scanner and 2D Image Fusion
2016
International audience; Map building becomes one of the most interesting research topic in computer vision field nowadays. To acquire accurate large 3D scene reconstructions, 3D laser scanners are recently developed and widely used. They produce accurate but sparse 3D point clouds of the environments. However, 3D reconstruction of rigidly moving objects along side with the large-scale 3D scene reconstruction is still lack of interest in many researches. To achieve a detailed object-level 3D reconstruction, a single scan of point cloud is insufficient due to their sparsity. For example, traditional Iterative Closest Point (ICP) registration technique or its variances are not accurate and rob…
Regularization operators for natural images based on nonlinear perception models.
2006
Image restoration requires some a priori knowledge of the solution. Some of the conventional regularization techniques are based on the estimation of the power spectrum density. Simple statistical models for spectral estimation just take into account second-order relations between the pixels of the image. However, natural images exhibit additional features, such as particular relationships between local Fourier or wavelet transform coefficients. Biological visual systems have evolved to capture these relations. We propose the use of this biological behavior to build regularization operators as an alternative to simple statistical models. The results suggest that if the penalty operator take…
Non-linear System Identification with Composite Relevance Vector Machines
2007
Nonlinear system identification based on relevance vector machines (RVMs) has been traditionally addressed by stacking the input and/or output regressors and then performing standard RVM regression. This letter introduces a full family of composite kernels in order to integrate the input and output information in the mapping function efficiently and hence generalize the standard approach. An improved trade-off between accuracy and sparsity is obtained in several benchmark problems. Also, the RVM yields confidence intervals for the predictions, and it is less sensitive to free parameter selection. Teoría de la Señal y Comunicaciones
Segmentation-Free Estimation of Aortic Diameters from MRI Using Deep Learning
2021
Accurate and reproducible measurements of the aortic diameters are crucial for the diagnosis of cardiovascular diseases and for therapeutic decision making. Currently, these measurements are manually performed by healthcare professionals, being time consuming, highly variable, and suffering from lack of reproducibility. In this work we propose a supervised deep-learning method for the direct estimation of aortic diameters. The approach is devised and tested over 100 magnetic resonance angiography scans without contrast agent. All data was expert-annotated at six aortic locations typically used in clinical practice. Our approach makes use of a 3D+2D convolutional neural network (CNN) that ta…
Contribution à l’apprentissage de représentation de données à base de graphes avec application à la catégorisation d’images
2020
Graph-based Manifold Learning algorithms are regarded as a powerful technique for feature extraction and dimensionality reduction in Pattern Recogniton, Computer Vision and Machine Learning fields. These algorithms utilize sample information contained in the item-item similarity and weighted matrix to reveal the intrinstic geometric structure of manifold. It exhibits the low dimensional structure in the high dimensional data. This motivates me to develop Graph-based Manifold Learning techniques on Pattern Recognition, specially, application to image categorization. The experimental datasets of thesis correspond to several categories of public image datasets such as face datasets, indoor and…
Event-related brain potentials of masked repetition and semantic priming while listening to sentences.
2012
We combined for the first time electrophysiological measures and masked priming technique in sentential context, by setting up a cross-modal masked priming paradigm involving the auditory presentation of sentences. ERPs were time-locked to an auditorily presented word that was preceded by a repeated, related or unrelated pattern masked prime. We registered a two-way N400-difference between unrelated and related/repeated primes, followed by a late positive component (LPC) for repetition priming. Related primes appear to facilitate the lexical-semantic processing of the target to the same extent repeated primes do (equally attenuated N400). Repetition priming exerts additional demands (LPC), …
Are coffee and toffee served in a cup? Ortho-phonologically mediated associative priming.
2008
We report three masked associative priming experiments with the lexical decision task that explore whether the initial activation flow of a visually presented word activates the semantic representations of that word's orthographic/phonological neighbours. The predictions of cascades and serial/modular models of lexical processing differ widely in this respect. Using a masked priming paradigm (stimulus onset asynchrony, SOA = 50 ms), words preceded by ortho-phonologically mediated associated “neighbours” ( oveja–MIEL, the Spanish for sheep–HONEY; note that oveja is a phonological neighbour of abeja, the Spanish for bee) were recognized more rapidly than words preceded by an unrelated word p…