Ponències/Comunicacions de congressos
http://hdl.handle.net/2117/3883
2016-05-02T08:47:55ZThree dimensional adaptive Laplacian Pyramid image coding
http://hdl.handle.net/2117/84991
Three dimensional adaptive Laplacian Pyramid image coding
Sallent Ribes, Sebastián; Torres Urgell, Lluís; Gils, L.
2016-03-31T13:56:46ZSallent Ribes, SebastiánTorres Urgell, LluísGils, L.Vector quantization in image sequence coding
http://hdl.handle.net/2117/84987
Vector quantization in image sequence coding
Huguet, Jordi; Torres Urgell, Lluís
2016-03-31T13:40:20ZHuguet, JordiTorres Urgell, LluísTemporal automatic edge detection of echocardiographic images
http://hdl.handle.net/2117/84722
Temporal automatic edge detection of echocardiographic images
Torres Urgell, Lluís; Gasull Llampallas, Antoni
2016-03-18T15:21:50ZTorres Urgell, LluísGasull Llampallas, AntoniObtencion de un esqueleto morfologico sin puntos redundantes
http://hdl.handle.net/2117/84338
Obtencion de un esqueleto morfologico sin puntos redundantes
Marqués Acosta, Fernando; Gasull Llampallas, Antoni; Torres Urgell, Lluís
Skeleton is a powerful method to describe binary images. Although there are different possibilities to define a skeleton, is very common to use morphological skeleton because it is easy and fast for implementation purposes. However, the set of points representing a morphological skeleton is redundant. This paper describes a method for selecting the mínimum number of points that have to be kept in order to reconstruct the binary image. Sorne results show that more than 50% of points can be saved.
2016-03-14T16:17:18ZMarqués Acosta, FernandoGasull Llampallas, AntoniTorres Urgell, LluísSkeleton is a powerful method to describe binary images. Although there are different possibilities to define a skeleton, is very common to use morphological skeleton because it is easy and fast for implementation purposes. However, the set of points representing a morphological skeleton is redundant. This paper describes a method for selecting the mínimum number of points that have to be kept in order to reconstruct the binary image. Sorne results show that more than 50% of points can be saved.Análisis de no estacionariedades en la interpolación de imágenes. Aplicación a codificación piramidal y zoom
http://hdl.handle.net/2117/84072
Análisis de no estacionariedades en la interpolación de imágenes. Aplicación a codificación piramidal y zoom
Gasull Llampallas, Antoni; Marqués Acosta, Fernando; Torres Urgell, Lluís
In some applications of digital image processing the fact that an image is a non estacionary signal can cause some undesirable effects. That is the case of block effect in transform coding, or destructive effects on contours in interpolation or decimation images. This paper deals with interpolation image case carried out by a previous segmentation of the image in estacionary regions, which can be interpolated in a classical way. The results have been applicated for zooming purposes, and in multirresolution image decomposition in order to reduce the error in decimation-interpolation process.
2016-03-09T15:05:03ZGasull Llampallas, AntoniMarqués Acosta, FernandoTorres Urgell, LluísIn some applications of digital image processing the fact that an image is a non estacionary signal can cause some undesirable effects. That is the case of block effect in transform coding, or destructive effects on contours in interpolation or decimation images. This paper deals with interpolation image case carried out by a previous segmentation of the image in estacionary regions, which can be interpolated in a classical way. The results have been applicated for zooming purposes, and in multirresolution image decomposition in order to reduce the error in decimation-interpolation process.An adaptive pyramid image coding system
http://hdl.handle.net/2117/83908
An adaptive pyramid image coding system
Torres Urgell, Lluís; Sallent Ribes, Sebastián
The Laplacian Pyramid is a new and efficient method for image encoding. The method is of increasing interest as band-pass pyramids and multiresolution images are being used in other image processing applications. The Pyramid Coding system is a hierarchical structure well suited for progressive image transmission over low-speed channels and hierarchical image retrieving in com- puterized image storage. The goal if this paper is to present a new pyramidal image structure that uses algorithms defined on arbitrary non-rectangular sampling lattices. An adaptive technique is also developed that assigns to the pyramidal structure an optimum sampling lattice as a function of the frequency content of the original image.
2016-03-07T16:29:12ZTorres Urgell, LluísSallent Ribes, SebastiánThe Laplacian Pyramid is a new and efficient method for image encoding. The method is of increasing interest as band-pass pyramids and multiresolution images are being used in other image processing applications. The Pyramid Coding system is a hierarchical structure well suited for progressive image transmission over low-speed channels and hierarchical image retrieving in com- puterized image storage. The goal if this paper is to present a new pyramidal image structure that uses algorithms defined on arbitrary non-rectangular sampling lattices. An adaptive technique is also developed that assigns to the pyramidal structure an optimum sampling lattice as a function of the frequency content of the original image.Diseño de filtros de imagen con funciones de transferencia de dominio romboidal
http://hdl.handle.net/2117/83820
Diseño de filtros de imagen con funciones de transferencia de dominio romboidal
Casas Rius, Joan Ramon; Torres Urgell, Lluís
When orthogonal sampllng patterns are applled to digitize analogue images, some degree of spatial redundancy is lntroduced. Quincunx sampllng patterns allow to remove this redundant visual lnformation which consist of diagonal image frequencies. Such processing ls usually done in the HD-MAC coding system for HDTV lmages. In this paper we suggest one method to design the 20-decimation fllters needed in quincunx subsampiing processes. The advantages of the design based on frequency domaln analysis are also pointed out. Finally, we show the performance of one of this filters.
2016-03-04T12:55:01ZCasas Rius, Joan RamonTorres Urgell, LluísWhen orthogonal sampllng patterns are applled to digitize analogue images, some degree of spatial redundancy is lntroduced. Quincunx sampllng patterns allow to remove this redundant visual lnformation which consist of diagonal image frequencies. Such processing ls usually done in the HD-MAC coding system for HDTV lmages. In this paper we suggest one method to design the 20-decimation fllters needed in quincunx subsampiing processes. The advantages of the design based on frequency domaln analysis are also pointed out. Finally, we show the performance of one of this filters.Detección automática de contornos en imágenes ecocardiográficas
http://hdl.handle.net/2117/83711
Detección automática de contornos en imágenes ecocardiográficas
Gasull Llampallas, Antoni; Marqués Acosta, Fernando; Sallent Ribes, Sebastián; Torres Urgell, Lluís; Vidal Sans, Josep
2016-03-02T14:08:33ZGasull Llampallas, AntoniMarqués Acosta, FernandoSallent Ribes, SebastiánTorres Urgell, LluísVidal Sans, JosepA new adaptive method for image compression using Karhunen-Loeve transform
http://hdl.handle.net/2117/83627
A new adaptive method for image compression using Karhunen-Loeve transform
Torres Urgell, Lluís
The Karhunen-Loeve transformation (KLT) is an optimal method for encoding images in the MSE sense. Only one spatially adaptive method is known to be reported in the literature. Many other suboptimal encoding methods have been developed to avoid the problems encountered in the application of the KLT. Such methods give performance which is inferior to the KLT in both MSE and visual quality, although some of them are quite efficient for first order Markov processes. A spatial domain adaptive algorithm, which, besides giving more compression than its nonadaptive counterpart, solves the problem of ill-conditioned eigenvectors is presented. It is also shown that for a large class of images the same code set may be used, thus avoiding the computational problem normally associated with the Karhunen-Loeve encoding scheme.
2016-03-01T13:19:12ZTorres Urgell, LluísThe Karhunen-Loeve transformation (KLT) is an optimal method for encoding images in the MSE sense. Only one spatially adaptive method is known to be reported in the literature. Many other suboptimal encoding methods have been developed to avoid the problems encountered in the application of the KLT. Such methods give performance which is inferior to the KLT in both MSE and visual quality, although some of them are quite efficient for first order Markov processes. A spatial domain adaptive algorithm, which, besides giving more compression than its nonadaptive counterpart, solves the problem of ill-conditioned eigenvectors is presented. It is also shown that for a large class of images the same code set may be used, thus avoiding the computational problem normally associated with the Karhunen-Loeve encoding scheme.A non-rectangular pyramid coding system
http://hdl.handle.net/2117/83135
A non-rectangular pyramid coding system
Sallent Ribes, Sebastián; Torres Urgell, Lluís
Rectangular sampling lattices bave been extensively applied to image coding. Sampling is a fundamental orera.tion m image coding systems as a proper selection of the sampling lattice minimizes the physica support of the informalion affecting the overall performance of the system and increasing the compression ratio. Non-rectangular sampling lattices have been occasionaly used in specific image systems but their application to coding schemes has not been reported in the literature. In this paper we present a characterization of multidimensional systems whose input and output are defined on arbitrary sampling structures. We also introduce the system response that eases the analysis of the sampling structure conversions. In the second part of the paper this theory is aplied to Laplacian Pyramid Coding. The result is the modelization of the algoríthms used in Pyramid Coding. In the third part of the paper this model is used to describe an efficient adaptive Pyramid image structure defined on arbitrary non-rectangular sampling lattices.
2016-02-18T15:59:06ZSallent Ribes, SebastiánTorres Urgell, LluísRectangular sampling lattices bave been extensively applied to image coding. Sampling is a fundamental orera.tion m image coding systems as a proper selection of the sampling lattice minimizes the physica support of the informalion affecting the overall performance of the system and increasing the compression ratio. Non-rectangular sampling lattices have been occasionaly used in specific image systems but their application to coding schemes has not been reported in the literature. In this paper we present a characterization of multidimensional systems whose input and output are defined on arbitrary sampling structures. We also introduce the system response that eases the analysis of the sampling structure conversions. In the second part of the paper this theory is aplied to Laplacian Pyramid Coding. The result is the modelization of the algoríthms used in Pyramid Coding. In the third part of the paper this model is used to describe an efficient adaptive Pyramid image structure defined on arbitrary non-rectangular sampling lattices.