.svg)
GIUV2015-260
The ISP research group, http://isp.uv.es, has a long tradition in statistical analysis of data coming from imaging systems. These measurements depend on the properties of the scenes and the physics of the imaging process, and their relevance depends on the (natural or artificial) observer that will analyze the data. Our distinct approach to signal, image and vision processing combines machine learning theory with the understanding of the underlying physics and biological vision. Applications mainly focus on optical remote sensing and computational visual neuroscience. Empirical statistical inference, also known as machine learning, is a field of computer science interested in making predictions, and models from observations and sensory data. The information processing tools in machine learning are critical to understand the function of natural neural networks involved in biological vision, as well as to make inferences in complex dynamic network systems, such as the Earth biosphere, atmosphere, and ecosystems. Problems in Visual Neuroscience and in Remote Sensing based geosciences require similar mathematical tools. For example, both scientific fields face model inversion and...The ISP research group, http://isp.uv.es, has a long tradition in statistical analysis of data coming from imaging systems. These measurements depend on the properties of the scenes and the physics of the imaging process, and their relevance depends on the (natural or artificial) observer that will analyze the data. Our distinct approach to signal, image and vision processing combines machine learning theory with the understanding of the underlying physics and biological vision. Applications mainly focus on optical remote sensing and computational visual neuroscience. Empirical statistical inference, also known as machine learning, is a field of computer science interested in making predictions, and models from observations and sensory data. The information processing tools in machine learning are critical to understand the function of natural neural networks involved in biological vision, as well as to make inferences in complex dynamic network systems, such as the Earth biosphere, atmosphere, and ecosystems. Problems in Visual Neuroscience and in Remote Sensing based geosciences require similar mathematical tools. For example, both scientific fields face model inversion and model understanding problems. In both cases, one has a complex forward model that is difficult to invert (to extract information from) either because it is not analytically invertible (undetermined) or because the measurements (or responses) are noisy in nature. In Remote Sensing, the forward model is the imaging process given certain state conditions in the surface and atmosphere. In Visual Neuroscience, the forward model includes what is known in the neural pathway from the retina to the different regions of the visual cortex. Inversion of such models is key to make quantitative and meaningful inferences about the underlying system that generated the observed data. Beyond such quantitative assessment, a qualitative interpretation of the proposed models is mandatory as well. Qualitative understanding is more challenging than prediction, and causal inference from empirical data is the common playground both in geoscience and neuroscience. Simultaneous observations and recordings from a phenomenon lead to multidimensional signals that may display strong statistical correlation between the components. However, correlation is not enough to establish cause-effect relationships. This is key when analyzing activation and inhibition in the communication between different brain regions, and it is also of paramount relevance when studying the causes, effects and confounders of essential climate variables for detection and attribution in climate science. Finally, another parallelism is the analysis of big visual data: hyperspectral imagery acquired by current and upcoming satellite sensors pose a big-data information processing problem in similar ways to that in the visual brain. Adaptation, pattern recognition, inference and decision making in the brain may be quite inspiring for remote sensing image analysis. The group is therefore organized into a theoretical research branch (A) and a more applied research branch (B). The theoretical machine learning core tackles model inversion, interpretation, causal inference from empirical data and inclusion of physical constraints and prior knowledge in big visual data. The applied research lines are devoted to apply and adapt the theoretrical developments for remote sensing, geociences and visual neuroscience. For the sake of simplicity, we have grouped together these activities along five conceptual research lines: machine learning, visual neuroscience, image processing, remote sensing and big data processing.
[Llegir més][Ocultar]
[Llegir més][Ocultar]
- Desenvolupament d'algorismes d'aprenentatge automàtic
- Desenvolupament de models estadístics de neurociència visual
- Aplicacions en processament d'imatges
- Aplicacions en teledetecció i geociències
- Aprenentatge automàtic. Desarrollo de técnicas de aprendizaje estadístico: redes neuronales, modelos gráficos, máquinas kernel, técnicas de clasificación, regresión, agrupamiento y visualización (manifold learning), aprendizaje activo, semisupervisado, relacional, Bayesiano, estructurado, y causal.
- Neurociència visual. Desarrollo de modelos y técnicas para el procesado de datos en neurociencia visual: manifold learning, independización, técnicas estadísticas de codificación óptima, de gaussianización, aprendizaje en variedades, estimación e inversión de modelos, interpretabilidad y causalidad.
- Processament d'imatges. Técnicas de procesado digital de imágenes: métricas de distorsión perceptual; compresión; estimación del movimiento; restauración; representación del color, detección de cambios, clasificación y segmentación de imágenes
- Teledetecció i geociència. Aplicaciones en tratamiento de señales e imágenes de teledetección: estimación de parámetros biofísicos y variables de flujos, inversión de modelos, segmentación de imágenes, detección de cambios y anomalías, fusión de imágenes y multiresolución, restauración, causalidad y atribución, ranking.
- Processament de gran volum de dades. Procesado de grandes bases de datos e imágenes de alta resolución temporal, espacial y espectral. Nuestros colaboradores (ESA, NASA, EUMETSAT, Google, DigitalGlobe) proporcionan acceso a grandes volúmenes de datos a procesar en tiempo real mediante técnicas de paralelización, clusters, y algoritmos.
| Nom | Caràcter de la participació | Entitat | Descripció |
|---|---|---|---|
| Gustavo Camps Valls | Director-a | UVEG-Valencia | Catedràtic-a d'Universitat |
| Equip d'investigació | |||
| Julia Carmen Amoros López | Membre | UVEG-Valencia | Professor-a Contractat-da Doctor-a |
| Javier Calpe Maravilla | Membre | UVEG-Valencia | Titular d'Universitat |
| Roberto Fernández Morán | Membre | UVEG-Valencia | Professor-a Ajudant-a Doctor-a |
| Luis Gómez Chova | Membre | UVEG-Valencia | Catedràtic-a d'Universitat |
| Valero Laparra Pérez-Muelas | Membre | UVEG-Valencia | Professor-a Ajudant-a Doctor-a |
| Jesús Malo López | Membre | UVEG-Valencia | Catedràtic-a d'Universitat |
| Jorge Muñoz Mari | Membre | UVEG-Valencia | Titular d'Universitat |
| Adrián Pérez Suay | Membre | UVEG-Valencia | Professor-a Ajudant-a Doctor-a |
| María Piles Guillem | Membre | UVEG-Valencia | Investigador-a contractat-ada Ramón y Cajal |
| Ana Belén Ruescas Orient | Membre | UVEG-Valencia | Professor-a Contractat-da Doctor-a |
