A5065759704- Numerical Methods and Algorithms
- Distributed and Parallel Computing Systems
- Advanced Electron Microscopy Techniques and Applications
- Electron and X-Ray Spectroscopy Techniques
- Advanced Optimization Algorithms Research
- Optimization and Packing Problems
- Medical Imaging Techniques and Applications
- Remote-Sensing Image Classification
- Advanced Data Compression Techniques
- Parallel Computing and Optimization Techniques
- Vehicle Routing Optimization Methods
- Cell Image Analysis Techniques
- Metaheuristic Optimization Algorithms Research
- Medical Image Segmentation Techniques
- Matrix Theory and Algorithms
- Polynomial and algebraic computation
- Computational Geometry and Mesh Generation
- Advanced X-ray and CT Imaging
- Digital Image Processing Techniques
- Evolutionary Algorithms and Applications
- Facility Location and Emergency Management
- graph theory and CDMA systems
- Image and Signal Denoising Methods
- Optimization and Search Problems
- Remote Sensing and Land Use
Universidad de Málaga
2011-2020
Vicomtech
2019
Universitat de València
2019
Biogipuzkoa Health Research Institute
2019
University of Almería
2004-2013
Budapest University of Technology and Economics
2010-2013
Universitat Politècnica de València
2004-2006
Centro Nacional de Biotecnología
2005
University of Pennsylvania
1996
Consejo Superior de Investigaciones Científicas
1996
Anomaly detection is an important task for hyperspectral data exploitation. A standard approach anomaly in the literature method developed by Reed and Xiaoli, also called RX algorithm. variation of this algorithm consists applying same concept to a local sliding window centered around each image pixel. The computational cost very high it strongly increases its versions. However, current advances performance computing help reduce run-time these algorithms. So, RX, possible achieve processing...
Spectral unmixing is an important technique for hyperspectral data exploitation, in which a mixed spectral signature decomposed into collection of spectrally pure constituent spectra, called endmembers, and set correspondent fractions, or abundances, that indicate the proportion each endmember present mixture. Over last years, several algorithms have been developed automatic semiautomatic extraction. Some available approaches assume input contains at least one distinct material further...
A continuous location problem in which a firm wants to set up two or more new facilities competitive environment is considered. Other offering the same product service already exist area. Both locations and qualities of are be found so as maximize profit obtained by firm. This global optimization problem, with many local optima. In this paper we analyze several approaches solve it, namely, three multistart search heuristics, simulated annealing algorithm, variants an evolutionary algorithm....
Journal Article FastSpMM: An Efficient Library for Sparse Matrix Product on GPUs Get access Gloria Ortega, Ortega * 1Department of Informatics, University Almería, Agrifood Campus International Excellence (CeiA3) Ctra. Sacramento s/n, 04120 Spain *Corresponding author: gloriaortega@ual.es Search other works by this author on: Oxford Academic Google Scholar Francisco Vázquez, Vázquez Inmaculada García, García 2Department Computer Architecture, Málaga, 29071 Ester M. Garzón The Journal, Volume...
Anomaly detection is an important task for hyperspectral data exploitation. A standard approach anomaly in the literature method developed by Reed and Yu, also called RX algorithm. It implements Mahalanobis distance, which has been widely used imaging applications. variation of this algorithm, known as kernel (KRX), consists applying same concept to a sliding window centered around each image pixel. KRX computationally very expensive because, every pixel, covariance matrix its inverse be...
Anomaly detection is an important task for hyperspectral data exploitation. Although many algorithms have been developed this purpose in recent years, due to the large dimensionality of image data, fast anomaly remains a challenging task. In work, we exploit computational power commodity graphics processing units (GPUs) and multicore processors obtain implementations well-known algorithm by Reed Xiaoli (RX algorithm), local (LRX) variant, which basically consists applying same concept...
The focus of this paper is on the analysis and evaluation a type parallel strategies applied to algorithm Advanced Multidimensional Interval Global Optimization (AMIGO). We investigate two versions AMIGO, called Parallel AMIGO (PAMIGO) algorithm, Global-PAMIGO Local-PAMIGO. idea behind our study that in order exploit potential parallelism algorithms, researchers need adapt them target computer architectures. Our PAMIGO algorithms have been designed for shared memory architectures are based...