Search results for " data"
showing 10 items of 7516 documents
Machine Learning VS Transfer Learning - Smart Camera Implementation for Face Authentication
2018
The aim of this paper is to highlight differences between classical machine learning and transfer learning applied to low cost real-time face authentication. Furthermore, in an access control context, the size of biometric data should be minimized so it can be stored on a remote personal media. These constraints have led us to compare only lightest versions of these algorithms. Transfer learning applied on Mobilenet v1 raises to 85% of accuracy, for a 457Ko model, with 3680s and 1.43s for training and prediction tasks. In comparison, the fastest integrated method (Random Forest) shows accuracy up to 90% for a 7,9Ko model, with a fifth of a second to be trained and a hundred of microseconds …
Fuzzy Fusion in Multimodal Biometric Systems
2007
Multimodal authentication systems represent an emerging trend for information security. These systems could replace conventional mono-modal biometric methods using two or more features for robust biometric authentication tasks. They employ unique combinations of measurable physical characteristics: fingerprint, facial features, iris of the eye, voice print, hand geometry, vein patterns, and so on. Since these traits are hardly imitable by other persons, the aim of these multibiometric systems is to achieve a high reliability to determine or verify person's identity. In this paper a multimodal biometric system using two different fingerprints is proposed. The matching module integrates fuzzy…
Database Versioning 2.0, a Transparent SQL Approach Used in Quantitative Management and Decision Making
2015
Abstract Managerial decisions are based on accurate information and in today's time raw data is produced even with a stroke of a key. Regardless of the data creating process one needs to know how the information was extracted and which pool of data was used. One important factor is time therefore we need to structure it in layers of data history in such a way that it can be analyzed, (post)process, in order to be able to retrieve valuable information. The simplest way is to use a Database Management System (DBMS), but even with such a management system we face the issue of making it a self-contained database on each version of data added. Our proposed system, a continuation of previous work…
Preliminary results of the project A.I.D.A. (Auto Immunity: Diagnosis Assisted by computer)
2014
In this paper, are presented the preliminary results of the A.I.D.A. (Auto Immunity: Diagnosis Assisted by computer) project which is developed in the frame of the cross-border cooperation Italy-Tunisia. According to the main objectives of this project, a database of interpreted Indirect ImmunoFluorescence (IIF) images on HEp 2 cells is being collected thanks to the contribution of Italian and Tunisian experts involved in routine diagnosis of autoimmune diseases. Through exchanging images and double reporting; a Gold Standard database, containing around 1000 double reported IIF images with different patterns including negative tests, has been settled. This Gold Standard database has been us…
LogDet divergence-based metric learning with triplet constraints and its applications.
2014
How to select and weigh features has always been a difficult problem in many image processing and pattern recognition applications. A data-dependent distance measure can address this problem to a certain extent, and therefore an accurate and efficient metric learning becomes necessary. In this paper, we propose a LogDet divergence-based metric learning with triplet constraints (LDMLT) approach, which can learn Mahalanobis distance metric accurately and efficiently. First of all, we demonstrate the good properties of triplet constraints and apply it in LogDet divergence-based metric learning model. Then, to deal with high-dimensional data, we apply a compressed representation method to learn…
A Two-Dimensional Autoregressive Model for MIMO Wideband Mobile Radio Channels
2008
In this work, we propose the multichannel two- dimensional (2D) autoregressive (AR) model for multiple-input multiple-output (MIMO) wideband mobile wireless channels. The parameters of the proposed model can be estimated from the real- world measurement data. For this purpose, we suggest using a straightforward extension of the prediction error minimization (PEM) algorithm. We also address the problem of possible instability of the multichannel 2D AR model. A model stabilization procedure based on numerical optimization techniques is proposed. The performance of the multichannel 2D AR model has been evaluated based on the synthetic data generated using two different channel simulators.
Learning-Graph-Based Quantum Algorithm for k-distinctness
2012
We present a quantum algorithm solving the $k$-distinctness problem in $O(n^{1-2^{k-2}/(2^k-1)})$ queries with a bounded error. This improves the previous $O(n^{k/(k+1)})$-query algorithm by Ambainis. The construction uses a modified learning graph approach. Compared to the recent paper by Belovs and Lee arXiv:1108.3022, the algorithm doesn't require any prior information on the input, and the complexity analysis is much simpler. Additionally, we introduce an $O(\sqrt{n}\alpha^{1/6})$ algorithm for the graph collision problem where $\alpha$ is the independence number of the graph.
Glycosyl azides as building blocks in convergent syntheses of oligomeric lactosamine and Lewisx saccharides
1997
Abstract Oligosaccharides containing type 2 lactosamine repeating units, e.g. neo-lacto-octaose and trimeric Lewis x derivatives, are constructed using neo-lactosamine azide building blocks. The azido group provides a favorable protection of the anomeric position which is stable to versatile protecting group manipulations and glycosylation reactions. On the other hand, glycosyl azides can be converted into glycosyl fluorides via a 1,3-dipolar cycloaddition with di- tert -butyl-acetylenedicar☐ylate and subsequent treatment of the resulting N -glycosyl triazoles with hydrogen fluoride-pyridine complex. Activation of the lactosamine fluorides with Lewis acids affords the possibility to extend …
A new method of anomeric protection and activation based on the conversion of glycosyl azides into glycosyl fluorides
1993
Glycosyl azides provide reliable anomeric protection stable to conditions for hydrolytic removal of ester groups, for reductive opening or release of acetalic diol protection, for the introduction of ether-type protection, and for glycosylation processes. The utility of this anomeric protection is further enhanced as glycosyl azides may be converted into glycosyl fluorides, which can be activated for glycosylation reactions. To this end, glycosyl azides have been subjected to 1,3-dipolar cycloaddition with di-tert-butyl acetylenedicarboxylate. On treatment with hydrogen fluoride-pyridine complex the N-glycosyl triazole derivatives directly give glycosyl fluorides.
Crystal and molecular structure studies of (Z)-N-methyl-C-4-substituted phenyl nitrones by XRD, DFT, FTIR and NMR methods
2017
Abstract (Z)-N-methyl-C-4-substituted phenyl nitrones –O+N(Me)=C(H)R (Z-2a R = 4-ClC6H4, Z-2b R = 4-NO2C6H4, Z-2c R = 4-CH3OC6H4) were synthesized and characterized by elemental analyses, FTIR, 1H, 13C and DEPT-135 NMR spectroscopy and also by single crystal X-ray diffraction (in the case of Z-2a and Z-2b). The geometries of the nitrone molecules Z-2a, Z-2b and Z-2c and their E-isomers; (E)-N-methyl-C-4-chlorophenyl nitrone E-2a, (E)-N-methyl-C-4-nitrophenyl nitrone E-2b and (E)-N-methyl-C-4-methoxyphenyl nitrone E-2c were optimized using density functional theory (DFT) at the B3LYP/6-311++G(d,p) level of theory. The theoretical vibrational frequencies obtained by DFT calculations are in go…