Natural Image Statistics

RESUMEN Se presenta una metodología de modelización para la generación de MDE basada en el método de simulación secuencial directa (SSD). La ventaja fundamental de este método geoestadístico es que tiene en cuenta y reproduce los patrones de variabilidad especial de los datos medidos sobre el terreno, en términos de conservación de los estadísticos básicos (histogramas, variogramas o covarianzas). Además permite generar un número ilimitado de posibles modelizaciones de la realidad que conserven las características de los datos experimentales. A partir de estos "modelos posibles" es posible introducir el concepto de "incertidumbre" derivada de los resultados de la modelización. Otra ventaja adicional radica en que evita los problemas del suavizado introducido por los métodos de estimación (triangulación, inversa distancias o krigeaje) que tienen su repercusión en los modelos derivados (pendientes, orientaciones, …) fundamentales en diversas aplicaciones.

The contribution of this work is the creation and development of a novel procedure for signals estimation, procedure LDM-G. The procedure will be developed with details and will consist of algorithms implementation for signals estimation via wavelet transform, applying the minimum description length principle for simultaneous and automatic selection of wavelet base and threshold value. By means of simulations, the implemented procedure was compared with techniques of universal threshold and Stein's SURE, where the effectiveness of the elaborated procedure was verified by means of the estimation of mean square error.

Feature representation has been widely used and developed recently. Multiscale features have led to remarkable breakthroughs for representation learning process in many computer vision tasks. This paper aims to provide a comprehensive survey of the recent multiscale representation learning achievements in classification tasks. Multiscale representation learning methods can be divided into two broad categories (multiscale geometric analysis and multiscale networks). Eleven kinds of multiscale geometric tools and seven kinds of multiscale networks are introduced. Some corresponding fundamental subproblems of these two broad categories are also described, including some concepts in representation process, specific representation methods with multiscale geometric analysis, and multiscale representation design strategies for networks. Then, the correlation between these two broad categories is illustrated, including respective characteristics, combination strategies, and characteristics of optimal representation. Some datasets and evaluation results are included to verify the effectiveness of the multiscale representation learning. Eventually, conclusion and future work are given, covering four directions [a) Choice and fusion; b) Self-adaption; c) Structure; and d) Generalization and proof]. Impact Statement-Multiscale representation learning techniques are fully utilized in existing artificial intelligence tasks. Compared to the existing papers, which pay attention to a particular task or algorithm, this paper is the first systematical review of multiscale

I also thank all administrators, technicians, and secretaries in the laboratory of LIMSI-CNRS which is associated to Paris Sud and Sorbonne Universities for their supports and help. We compare our methods with the state-of-the-art algorithms using a standard database of robot localization. We study the influence of system parameters and compare different conditions on the same dataset. These experiments show that our proposed model, while being very simple, leads to state-of-the-art results on a semantic place recognition task.

This paper proposes an alternative approach for the problem of Arabic handwritten character recognition. The proposed model is based on Deep Belief Networks (DBNs) which are unsupervised machine learning methods. A greedy layer-wise fashion based on Restricted Boltzmann Machines and contrastive divergence learning algorithm will be used to train such model. Previous studies have shown that DBNs are capable to extract a set of sparse features, which can be used to code the initial data in an efficient way. The assumption is that such representation must improve the linear separation among the different classes and thus a simple classification algorithm, like softmax regression, should be sufficient to achieve accurate recognition rates. The literature reviewed showed that this alternative approach has not been considered yet in the context of Arabic character recognition, which deserves to be investigated and evaluate its performance for such problem.

I also thank all administrators, technicians, and secretaries in the laboratory of LIMSI-CNRS which is associated to Paris Sud and Sorbonne Universities for their supports and help. We compare our methods with the state-of-the-art algorithms using a standard database of robot localization. We study the influence of system parameters and compare different conditions on the same dataset. These experiments show that our proposed model, while being very simple, leads to state-of-the-art results on a semantic place recognition task.

Determining the statistical relationships of images that facilitate robust visual behaviour is nontrivial. Here we ask if some spatial relationships are more easily learned by the visual brain than others. Visually naïve bumblebees were trained to recognise coloured artificial flowers in scenes of equal spatial complexity but differing patterns of stimulus intensity. When flowers of similar intensity were grouped into extended regions across the array (coincident with natural patterns of light), the accuracy of the bees' foraging behaviour was dependent on spatial context, even though this information was redundant to the task. When the same intensity information was organised into a pattern that was less consistent with natural patterns of illumination but of equal order, their behaviour was independent of spatial context and they required double the training time to solve the same conditional task. These observations suggest the brain is biased to more efficiently encode/learn ecologically 'meaningful' image correlations.

A successful class of image denoising methods is based on Bayesian approaches working in wa- velet representations. The performance of these methods improves when relations among the local frequency coefficients are explicitly included . However, in these techniques, analytical estimates can be obtained only for particular combinations of analytical models of signal and noise, thus precluding its straightforward extension to

A successful class of image denoising methods is based on Bayesian approaches working in wa- velet representations. The performance of these methods improves when relations among the local frequency coefficients are explicitly included . However, in these techniques, analytical estimates can be obtained only for particular combinations of analytical models of signal and noise, thus precluding its straightforward extension to

Las técnicas de reducción de ruido son ampliamente utilizadas en la grabación de audio, la edición de imágenes y en el procesamiento de señales industriales. La idea es reconstruir los datos originales a partir de la señal ruidosa suprimiendo toda, o casi toda, la distorsión generada por el ruido inherente a los procesos físicos. En el presente trabajo se realiza una comparación de diferentes métodos de supresión de ruido que se basan en la selección adaptativa del umbral. Estas técnicas han sido usadas extensivamente en el procesamiento de imágenes pero el objetivo de este trabajo es evaluar su rendimiento en la reducción de ruido de señales industriales. En particular se analiza el comportamiento de los métodos Bayes Shrink, Normal Shrink, Modified Shrink y Neight Shrink para la reducción de ruido gaussiano en estas señales. A tales efectos se utilizó un conjunto de señales patrón, que incluye a las señales propuestas por Donoho y otras mediciones representativas obtenidas de procesos reales en las plantas de Níquel cubanas. Las pruebas realizadas revelan que el algoritmo Neigh Shrink es el que mejor se comporta en los datos analizados.

A successful class of image denoising methods is based on Bayesian approaches working in wa- velet representations. The performance of these methods improves when relations among the local frequency coefficients are explicitly included . However, in these techniques, analytical estimates can be obtained only for particular combinations of analytical models of signal and noise, thus precluding its straightforward extension to

Las técnicas de reducción de ruido son ampliamente utilizadas en la grabación de audio, la edición de imágenes y en el procesamiento de señales industriales. La idea es reconstruir los datos originales a partir de la señal ruidosa suprimiendo toda, o casi toda, la distorsión generada por el ruido inherente a los procesos físicos. En el presente trabajo se realiza una comparación de diferentes métodos de supresión de ruido que se basan en la selección adaptativa del umbral. Estas técnicas han sido usadas extensivamente en el procesamiento de imágenes pero el objetivo de este trabajo es evaluar su rendimiento en la reducción de ruido de señales industriales. En particular se analiza el comportamiento de los métodos Bayes Shrink, Normal Shrink, Modified Shrink y Neight Shrink para la reducción de ruido gaussiano en estas señales. A tales efectos se utilizó un conjunto de señales patrón, que incluye a las señales propuestas por Donoho y otras mediciones representativas obtenidas de procesos reales en las plantas de Níquel cubanas. Las pruebas realizadas revelan que el algoritmo Neigh Shrink es el que mejor se comporta en los datos analizados.

The object of radiation therapy as a therapeutic technique in oncology is introduced. In order to have success in the treatment of a tumor, one needs to characterize the dose (energy absorbed in mass) delivered to a patient. The knowledge of the interaction mechanisms between radiation and matter is the basis for radiation dosimetry. The properties of the dose deposition in matter due to different types of radiation determine their applicability for cancer therapy. A summary of different kinds and sources of radiation is presented first, including natural processes that can generate radiation inside an absorbing medium. The aim is to present the dose deposition inside a patient as a complex procedure, where all these phenomena need to be properly characterized in order to calculate the absorbed dose in a medium. We focus on photons and electrons, which are the main types of radiation employed in cancer therapy. A photon beam transfers its energy to electrons, which finally deposit the dose in a medium. We describe in detail the interaction mechanisms of photons and electrons with matter. These phenomena give us the foundations for the absorbed dose calculation and measurement.

La radioterapia es el tratamiento más aplicado a los pacientes que padecen cáncer, donde es irradiado el volumen del tumor, tomando en cuenta diversos criterios para evitar dañar los tejidos sanos, por tanto, es importante tener la certeza de que la dosis suministrada es la correcta. Con el fin de desarrollar un método precisó de verificación del planeamiento de tratamiento se construyo objetos simuladores de pulmón y tejido blando a base de resina epoxi, esferas fenolíticas y carbonato de calcio, estos fueron analizados por microscopia de barrido electrónico y validados mediante simulación Monte Carlo usando código Penélope v.2008 en comparación con publicaciones del Instituto Nacional de Estándares y Tecnología (NIST por sus siglas en ingles). Con esto se evaluó la interacción de la radiación con objetos simuladores homogéneos (pulmón) y heterogéneos (pulmón, tejido blando, pulmón) construidos y simulados. Siendo el principal objetivo de este trabajo el uso de los objetos simulado...

Publicación anticipada El comité editorial de la revista Ingeniería y Competitividad informa que este artículo fue aprobado para publicación en el volumen 25 número 1 teniendo en cuenta los requisitos editoriales y los conceptos emitidos por los pares evaluadores. Por lo tanto, se publica anticipadamente para su consulta, descarga y citación provisional, aclarando que esta puede diferir de la versión final, ya que no ha completado las etapas finales del proceso editorial (corrección de estilo, traducción y diagramación) y solo los títulos, datos de autores, palabras clave y resúmenes corresponden a la versión final del artículo. Como citar: Wilches-Visbal JH, Apaza-Veliz D. Validación de la reconstrucción espectral de un haz de electrones por comparación con espectro de Monte Carlo. INGENIERÍA Y COMPETITIVIDAD, In press 2023.

The visual world is complex and continuously changing. Yet, our brain transforms patterns of light falling on our retina into a coherent percept within a few hundred milliseconds. Possibly, low-level neural responses already carry substantial information to facilitate rapid characterization of the visual input. Here, we computationally estimated low-level contrast responses to computer-generated naturalistic images, and tested whether spatial pooling of these responses could predict image similarity at the neural and behavioral level. Using EEG, we show that statistics derived from pooled responses explain a large amount of variance between single-image evoked potentials (ERPs) in individual subjects. Dissimilarity analysis on multi-electrode ERPs demonstrated that large differences between images in pooled response statistics are predictive of more dissimilar patterns of evoked activity, whereas images with little difference in statistics give rise to highly similar evoked activity patterns. In a separate behavioral experiment, images with large differences in statistics were judged as different categories, whereas images with little differences were confused. These findings suggest that statistics derived from low-level contrast responses can be extracted in early visual processing and can be relevant for rapid judgment of visual similarity. We compared our results with two other, well-known contrast statistics: Fourier power spectra and higherorder properties of contrast distributions (skewness and kurtosis). Interestingly, whereas these statistics allow for accurate image categorization, they do not predict ERP response patterns or behavioral categorization confusions. These converging computational, neural and behavioral results suggest that statistics of pooled contrast responses contain information that corresponds with perceived visual similarity in a rapid, low-level categorization task.

Activation functions (AF) are necessary components of neural networks that allow approximation of functions, but AFs in current use are usually simple monotonically increasing functions. In this paper, we propose trainable compound AF (TCA) composed of a sum of shifted and scaled simple AFs. TCAs increase the effectiveness of networks with fewer parameters compared to added layers. TCAs have a special interpretation in generative networks because they effectively estimate the marginal distributions of each dimension of the data using a mixture distribution, reducing modality and making linear dimension reduction more effective. When used in restricted Boltzmann machines (RBMs), they result in a novel type of RBM with mixture-based stochastic units. Improved performance is demonstrated in experiments using RBMs, deep belief networks (DBN), projected belief networks (PBN), and variational autoencoders (VAE).

Student theses are made available in the TU/e repository upon obtaining the required degree. The grade received is not published on the document as presented in the repository. The required complexity or quality of research of student theses may vary by program, and the required minimum study period may vary in duration. General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain TU/e Eindhoven University of Technology Image reconstruction from multiscale top points Master of Science thesis

Restricted Boltzmann Machine (RBM) is an energy-based, undirected graphical model. It is commonly used for unsupervised and supervised machine learning. Typically, RBM is trained using contrastive divergence (CD). However, training with CD is slow and does not estimate the exact gradient of the log-likelihood cost function. In this work, the model expectation of gradient learning for RBM has been calculated using a quantum annealer (D-Wave 2000Q), where obtaining samples is faster than Markov chain Monte Carlo (MCMC) used in CD. Training and classification results of RBM trained using quantum annealing are compared with the CD-based method. The performance of the two approaches is compared with respect to the classification accuracies, image reconstruction, and log-likelihood results. The classification accuracy results indicate comparable performances of the two methods. Image reconstruction and log-likelihood results show improved performance of the CD-based method. It is shown th...

In computer vision, measurement of image properties such as color or texture is essential. In this paper, we propose a solid framework for the local measurement of texture in color images. We give a physical basis for the integration of the well-known Gabor filters with the measurement of color. Our method implies that the color-texture is measured in the wavelength-Fourier domain. The measurement filter in this domain boils down to a 3D Gaussian, representing a Gabor-Gaussian in the spatial-color domain. In addition, the extended measurements invariant to shadow or shading can be derived directly from the proposed method. We apply a simple segmentation algorithm to illustrate the performance of our proposed color-texture measurement. The derived color-texture filter is demonstrated to be accurate in capturing texture statistics. In addition, the method is compatible with the measurement of gray-value texture.

Binocular fusion relies on matching points in the two eyes that correspond to the same physical feature in the world; however, not all world features are binocularly visible. Near depth edges, some regions of a scene are often visible to only one eye (so-called half occlusions). Accurate detection of these monocularly visible regions is likely to be important for stable visual perception. If monocular regions are not detected as such, the visual system may attempt to binocularly fuse non-corresponding points, which can result in unstable percepts. We investigated the hypothesis that the visual system capitalizes on statistical regularities associated with depth edges in natural scenes to aid binocular fusion and facilitate perceptual stability. By sampling from a large set of stereoscopic natural images with co-registered distance information, we found evidence that monocularly visible regions near depth edges primarily result from background occlusions. Accordingly, monocular regions tended to be more visually similar to the adjacent binocularly visible background region than to the adjacent binocularly visible foreground. Consistent with our hypothesis, perceptual experiments showed that perception tended to be more stable when the image properties of the depth edge were statistically more likely given the probability of occurrence in natural scenes (i.e., when monocular regions were more visually similar to the binocular background). The generality of these results was supported by a parametric study with simulated environments. Exploiting regularities in natural environments may allow the visual system to facilitate fusion and perceptual stability when both binocular and monocular regions are visible. world, the visual system exploits statistical regularities that are present in binocular images of natural scenes

El filtrado de señales de voz con el objeto de eliminar en lo posible los ruidos producidos por el sistema de grabación utilizado constituye una etapa previa en todo sistema de análisis del habla. En este documento se describen las experiencias que se están realizando en el filtrado de señales de voz utilizando transformadas wavelet ([5]). Se han desarrollado un conjunto de módulos implementados en Matlab [11] que permiten filtrar una señal de voz, apoyándose en las funciones de cálculo en el dominio wavelet (UviWave) realizadas por el

Data compression is the process of encoding information using fewer bits than the original representation. The main idea of compression is to eliminate the redundancy in the original representation. Data Compression technique is devided into two namely: lossy compression and lossless compression. This paper is studying lossy image compression based on three transformations: Curvelet, Wavelet and Contourlet of grayscale images. In this research, a comparison is presented for the digital image compression among three transformations by converting a color digital image of size 512*512 to a gray scale digital image then decomposing it to its coefficients at various levels, then reconstructing the image after neglecting the effect of some of its high frequencies to evaluate how much the compressed image is closed to the original image. Practical results clarify that Curvelet Transform is better than Wavelet while Contourlet is much better in image compression. Evaluation metrices are: Correlation, MSE, PSNR and CR. The numerical results have been achieved using MATLAB software.

Deformational plagiocephaly is a term describing cranial asymmetry and deformation commonly seen in infants. The purpose of this work was to develop a methodology for assessment and modelling of head asymmetry. The clinical population consisted of 38 infants for whom 3-dimensional surface scans of the head had been obtained both before and after their helmet orthotic treatment. Non-rigid registration of a symmetric template to each of the scans provided detailed point correspondence between scans. A new asymmetry measure was defined and was used in order to quantify and localize the asymmetry of each infant's head, and again employed to estimate the improvement of asymmetry after the helmet therapy. A statistical model of head asymmetry was developed (PCA). The main modes of variation were in good agreement with clinical observations, and the model provided an excellent and instructive quantitative description of the asymmetry present in the dataset.

análisis de reduCCión de ruido en señales eeg orientado al reConoCimiento de patrones cristian guarnizo LEmus 1 Resumen En este artículo se presenta un estudio sobre la reducción de ruido de fondo (electrónico) en señales electroencefalográficas (EEG) utilizando la transformada wavelet, asumiendo que las características extraídas son susceptibles al ruido inherente en la señal y común entre clases. Adicionalmente se observa la incidencia en la separación de las muestras en el espacio de características a partir de un clasificador bayesiano lineal. Se obtiene un incremento del 1% en la media del porcentaje de acierto, al realizar la reducción de ruido en la identificación de dos estados funcionales.

Evidence that is wrongly taken to show that perception is conceptual I have mentioned two kinds of evidence for the nonconceptuality of perception: The first-evidence that I did not discuss in detail-involved the finegrained nature of perceptual representation as compared with conceptual representation (Evans, 1982). The second involved the visual abilities of infants, insects, and wasps without evidence of conceptual abilities that are linked to the visual abilities (Burge, 2010a; Peacocke, 2001a). There is also evidence that has been taken to support conceptual representation in perception. The purpose of this chapter is negative-to rebut this evidence.

The Quincunx wavelet, the lifting scheme wavelet and the second generation bandelet transform are a new method to offer an optimal representation for image geometric. In the field of medical diagnostics, interested parties have resorted increasingly to medical imaging. It is well established that the accuracy and completeness of diagnosis are initially connected with the image quality, but the quality of the image is itself dependent on a number of factors including primarily the processing that an image must undergo to enhance its quality. The quality evaluation of compressed image is necessary to judge the performance of a compression method. This paper introduces an algorithm for medical image compression based on hybrid non-subsampled contourlet (NSCT) and quincunx wavelet transforms (QWT) coupled with SPIHT coding algorithm, of which we present the objective measurements (PSNR, EDGE, WPSNR, MSSIM, VIF, and WSNR) in order to evaluate the quality of the image.

The Quincunx wavelet, the lifting scheme wavelet and the second generation bandelet transform are a new method to offer an optimal representation for image geometric. In the field of medical diagnostics, interested parties have resorted increasingly to medical imaging. It is well established that the accuracy and completeness of diagnosis are initially connected with the image quality, but the quality of the image is itself dependent on a number of factors including primarily the processing that an image must undergo to enhance its quality. The quality evaluation of compressed image is necessary to judge the performance of a compression method. This paper introduces an algorithm for medical image compression based on hybrid non-subsampled contourlet (NSCT) and quincunx wavelet transforms (QWT) coupled with SPIHT coding algorithm, of which we present the objective measurements (PSNR, EDGE, WPSNR, MSSIM, VIF, and WSNR) in order to evaluate the quality of the image.

Para usar la información sensorial de manera eficiente para emitir juicios y guiar la acción en el mundo, el cerebro debe representar y usar información sobre la incertidumbre en sus cálculos para la percepción y la acción. Los métodos bayesianos han demostrado ser exitosos en la construcción de teorías computacionales para la percepción y el control sensoriomotor, y la psicofísica está proporcionando un creciente cuerpo de evidencia de que los cálculos perceptivos humanos son ``óptimos de Bayes''. Esto lleva a la ``hipótesis de la codificación bayesiana'': que el cerebro representa la información sensorial de manera probabilística, en forma de distribuciones de probabilidad. Recientemente se han propuesto varios esquemas computacionales sobre cómo se podría lograr esto en poblaciones de neuronas. Los datos neurofisiológicos sobre la hipótesis, sin embargo, son casi inexistentes. Un gran desafío para los neurocientíficos es probar estas ideas experimentalmente y así determinar si las neuronas codifican información sobre la incertidumbre sensorial y cómo lo hacen.

Speckle noise in ultrasound (US) medical images is the prime factor that undermines its full utilization. This noise is added by the constructive / destructive interference of sound waves travelling through hard-and soft-tissues of a patient. It is therefore generally accepted that the noise is unavoidable. As an alternate researchers have proposed several algorithms to somewhat undermine the effect of speckle noise. The discrete wavelet transform (DWT) has been used by several researchers. However, the performance of only a few transforms has been demonstrated. This paper provides a comparison of several DWT. The algorithm comprises of a pre-processing stage using Wiener filter, and a post-processing stage using Median filter. The processed image is compared with the original image on four metrics: two are based on full-reference (FR) image quality assessment (IQA), and the remaining two are based on noreference (NR) IQA metrics. The FR-IQA are peak signal-tonoise ratio (PSNR) and mean structurally similarity index measure (MSSIM). The two NR-IQA techniques are blind pseudo-reference image (BPRI), and blind multiple pseudoreference images (BMPRI). It has been demonstrated that some of these wavelet transforms outperform others by a significant margin.

Sparse representations have been often used for inverse problems in signal and image processing. Furthermore, frameworks for signal classification using sparse and overcomplete representations have been developed. Data-dependent representations using learned dictionaries have been significant in applications such as feature extraction and denoising. In this paper, our goal is to perform pattern classification in a domain referred to as the data representation domain, where data from different classes are sparsely ...

El objetivo de este trabajo es el estudio de un método de pre-procesamiento de los datos medidos por un tomógrafo optoacústico bidimensional para reducir o eliminar los artefactos introducidos por la escasa cantidad de detectores en el sistema experimental y el acotado ancho de banda de estos. Para esta tarea, se utilizó una red neuronal profunda generativa adversaria y se comparó su rendimiento con una red neuronal de referencia U-Net. En la mayoría de los casos de testeo realizados, se encontró una leve mejora aplicando la red propuesta al medir la correlación de Pearson y la relación señal a ruido piso entre la imagen reconstruida producto de los datos procesados por el modelo y la imagen de alta resolución de referencia.

Picture quality Measures play main role in the picture processing of a kind. Main objective is to evaluate quality that applies to estimate operations and image processing techniques. In recent years, significant effort has been made to advance objective that links fit with objective human quality measures or subjective styles. In addition, results from the full reference (FR) approach, which was based on pixel-to-pixel error.

In computer vision, measurement of image properties such as color or texture is essential. In this paper, we propose a solid framework for the local measurement of texture in color images. We give a physical basis for the integration of the well-known Gabor filters with the measurement of color. Our method implies that the color-texture is measured in the wavelength-Fourier domain. The measurement filter in this domain boils down to a 3D Gaussian, representing a Gabor-Gaussian in the spatial-color domain. In addition, the extended measurements invariant to shadow or shading can be derived directly from the proposed method. We apply a simple segmentation algorithm to illustrate the performance of our proposed color-texture measurement. The derived color-texture filter is demonstrated to be accurate in capturing texture statistics. In addition, the method is compatible with the measurement of gray-value texture.

In this paper we propose to make Bayesian inferences for the parameters of the Lomax distribution using non-informative priors, namely the Jeffreys prior and the reference prior. We assess Bayesian estimation through a Monte Carlo study with 500 simulated data sets. To evaluate the possible impact of prior specification on estimation, two criteria were considered: the bias and square root of the mean square error. The developed procedures are illustrated on a real data set.

Con el desarrollo de la tecnología, los objetivos principales para toda transmisión realizada, por un canal de comunicaciones, han sido optimizar el ancho de banda, mantener la calidad y eficiencia de la transmisión; por lo que, dentro del tema de estudio, la compresión de imágenes juega un papel fundamental, ya que facilita la transmisión de la misma a lo largo del canal, mejorando la eficiencia del ancho de banda. El desarrollo de Televisión Digital en el Ecuador es un hecho, y su éxito depende de

Debido a la carencia de la aplicación de las ondeletas en datos geodésicos ya sea por su complejidad o por la carencia de especialistas en el tema, es que se decide aplicar las wavelets a datos de algunas estaciones continuas ubicadas en Colombia para estimar la velocidad de desplazamiento de esa zona en particular a partir de datos de una serie de tiempo superior a los dos años de funcionamiento de la estación.

Members of the equivalence class of convex Lambertian surfaces that produce the same set of orthographic images under arbitrary combinations of distant point light sources are related by elements of a three-parameter subgroup of GL.3/, called generalized bas-relief (GBR) transformations. This inherent ambiguity in determining the three-dimensional shape of an object from shading and shadow information is called the bas-relief ambiguity. K. Ikeuchi (ed.

Global support radial base functions are effective in eliminating impulsive noise in non-linear processing. This paper introduces an interpolation technique which efficiently reduces image impulsive noise by means of an interpolant obtained through radial base functions. These functions have been used in a research project designed to develop a system for the recovery of images of Amazonian aquatic resources. This technique starts with the tagging by interpolation of noisy image pixels. Thus, a value of reconstruction for the noisy pixels is generated using neighboring pixels. The results obtained with this technique have proved comparable and often better than those obtained with previously known techniques. According to results analysis, this technique can be successfully applied on images with high noise levels. The results are low error in noisy pixel reconstruction and better visual quality.