|
|
Publications about Texture
Result of the query in the list of publications :
2 Articles |
1 - Geometric Feature Extraction by a Multi-Marked Point Process .
F. Lafarge and G. Gimel'farb and X. Descombes. IEEE Trans. Pattern Analysis and Machine Intelligence, 32(9): pages 1597-1609, September 2010.
Keywords : Shape extraction, Spatial point process, Stochastic geometry, fast optimization, Texture, remote sensing.
@ARTICLE{pami09b_lafarge,
|
| author |
= |
{Lafarge, F. and Gimel'farb, G. and Descombes, X.}, |
| title |
= |
{Geometric Feature Extraction by a Multi-Marked Point Process }, |
| year |
= |
{2010}, |
| month |
= |
{September}, |
| journal |
= |
{IEEE Trans. Pattern Analysis and Machine Intelligence}, |
| volume |
= |
{32}, |
| number |
= |
{9}, |
| pages |
= |
{1597-1609}, |
| url |
= |
{http://dx.doi.org/10.1109/TPAMI.2009.152}, |
| keyword |
= |
{Shape extraction, Spatial point process, Stochastic geometry, fast optimization, Texture, remote sensing} |
| } |
Abstract :
This paper presents a new stochastic marked point process for describing images in terms of a finite library of geometric objects. Image analysis based on conventional marked point processes has already produced convincing results but at the expense of parameter tuning, computing time, and model specificity. Our more general multimarked point process has simpler parametric setting, yields notably shorter computing times, and can be applied to a variety of applications. Both linear and areal primitives extracted from a library of geometric objects are matched to a given image using a probabilistic Gibbs model, and a Jump-Diffusion process is performed to search for the optimal object configuration. Experiments with remotely sensed images and natural textures show that the proposed approach has good potential. We conclude with a discussion about the insertion of more complex object interactions in the model by studying the compromise between model complexity and efficiency. |
|
2 - A study of Gaussian mixture models of colour and texture features for image classification and segmentation.
H. Permuter and J.M. Francos and I. H. Jermyn. Pattern Recognition, 39(4): pages 695--706, April 2006.
Keywords : Classification, Segmentation, Texture, Colour, Gaussian mixture, Decison fusion.
@ARTICLE{permuter_pr06,
|
| author |
= |
{Permuter, H. and Francos, J.M. and Jermyn, I. H.}, |
| title |
= |
{A study of Gaussian mixture models of colour and texture features for image classification and segmentation}, |
| year |
= |
{2006}, |
| month |
= |
{April}, |
| journal |
= |
{Pattern Recognition}, |
| volume |
= |
{39}, |
| number |
= |
{4}, |
| pages |
= |
{695--706}, |
| url |
= |
{http://dx.doi.org/10.1016/j.patcog.2005.10.028}, |
| pdf |
= |
{ftp://ftp-sop.inria.fr/ariana/Articles/2006_permuter_pr06.pdf}, |
| keyword |
= |
{Classification, Segmentation, Texture, Colour, Gaussian mixture, Decison fusion} |
| } |
Abstract :
The aims of this paper are two-fold: to define Gaussian mixture models of coloured texture on several feature paces and to compare the performance of these models
in various classification tasks, both with each other and with other models popular in the literature. We construct Gaussian mixtures models over a variety of different colour and texture feature spaces, with a view to the retrieval of textured colour images from databases. We compare supervised classification results for different choices of colour and texture features using the Vistex database, and explore the best set of features and the best GMM configuration for this task. In addition we introduce several methods for combining the 'colour' and 'structure' information in order to improve the classification performance. We then apply the resulting models to the classification of texture databases and to the classification of man-made and natural areas in aerial images. We compare the GMM model with other models in the literature, and show an overall improvement in performance. |
|
 top of the page
2 PhD Thesis and Habilitations |
1 - Analyse de texture dans l'espace hyperspectral par des méthodes probabilistes.
G. Rellier. PhD Thesis, Universite de Nice Sophia Antipolis, November 2002.
Keywords : Hyperspectral imaging, Texture, Classification, Markov Fields.
@PHDTHESIS{rellier,
|
| author |
= |
{Rellier, G.}, |
| title |
= |
{Analyse de texture dans l'espace hyperspectral par des méthodes probabilistes}, |
| year |
= |
{2002}, |
| month |
= |
{November}, |
| school |
= |
{Universite de Nice Sophia Antipolis}, |
| url |
= |
{https://hal.inria.fr/tel-00505898}, |
| keyword |
= |
{Hyperspectral imaging, Texture, Classification, Markov Fields} |
| } |
Résumé :
Dans cette thèse, on aborde le problème de l'analyse de texture pour l'étude des zones urbaines. La texture est une notion spatiale désignant ce qui, en dehors de la couleur ou du niveau de gris, caractérise l'homogénéité visuelle d'une zone donnée d'une image. Le but de cette étude est d'établir un modèle qui permette une analyse de texture prenant en compte conjointement l'aspect spatial et l'aspect spectral, à partir d'images hyperspectrales. Ces images sont caractérisées par un nombre de canaux largement supérieur à celui des images multispectrales classiques. On désire tirer parti de l'information spectrale pour améliorer l'analyse spatiale. Les textures sont modélisées par un champ de Markov gaussien vectoriel, qui permet de prendre en compte les relations spatiales entre pixels, mais aussi les relations inter-bandes à l'intérieur d'un même pixel. Ce champ est adapté aux images hyperspectrales par une simplification évitant l'apparition de problèmes d'estimation statistique dans des espaces de grande dimension. Dans le but d'éviter ces problèmes, on effectue également une réduction de dimension des données grâce à un algorithme de poursuite de projection. Cet algorithme permet de déterminer un sous-espace de projection dans lequel une grandeur appelée indice de projection est optimisée. L'indice de projection est défini par rapport à la modélisation de texture proposée, de manière à ce que le sous-espace optimal maximise la distance entre les classes prédéfinies, dans le cadre de la classification. La méthode d'analyse de texture est testée dans le cadre d'une classification supervisée. Pour ce faire, on met au point deux algorithmes que l'on compare avec des algorithmes classiques utilisant ou non l'information de texture. Des tests sont réalisés sur des images hyperspectrales AVIRIS. |
Abstract :
In this work, we investigate the problem of texture analysis of urban areas. Texture is a spatial concept that refers to the visual homogeneity characteristics of an image, not taking into account color or grey level. The aim of this research is to define a model which allows a joint spectral and spatial analysis of texture, and then to apply this model to hyperspectral images. These images many more bands than classical multispectral images. We intend to make use of spectral information and improve simple spatial analysis. Textures are modeled by a vectorial Gauss-Markov random field, which allows us to take into account the spatial interactions between pixels as well as inter-band relationships for a single pixel. This field has been adapted to hyperspectral images by a simplification which avoids statistical estimation problems common to high dimensional spaces. In order to avoid these problems, we also reduce the dimensionality of the data, using a projection pursuit algorithm. This algorithm determines a projection subspace in which an index, called projection index, is optimized. This index is defined in relation to the proposed texture model so that, when a classification is being carried out, the optimal subspace maximizes the distance between predefined training samples. This texture analysis method is tested within a supervised classification framework. For this purpose, we propose two classification algorithms that we compare to two classical algorithms, one which uses texture information and one which does not. Tests are carried out on AVIRIS hyperspectral images. |
|
2 - Analyse de Texture par Méthodes Markoviennes et par Morphologie Mathématique : Application à l'Analyse des Zones Urbaines sur des Images Satellitales.
A. Lorette. PhD Thesis, Universite de Nice Sophia Antipolis, September 1999.
Keywords : Texture, Segmentation, Markov Fields, Mathematical morphology, Urban areas.
@PHDTHESIS{lorette99,
|
| author |
= |
{Lorette, A.}, |
| title |
= |
{Analyse de Texture par Méthodes Markoviennes et par Morphologie Mathématique : Application à l'Analyse des Zones Urbaines sur des Images Satellitales}, |
| year |
= |
{1999}, |
| month |
= |
{September}, |
| school |
= |
{Universite de Nice Sophia Antipolis}, |
| pdf |
= |
{Theses/these-lorette.pdf}, |
| keyword |
= |
{Texture, Segmentation, Markov Fields, Mathematical morphology, Urban areas} |
| } |
Résumé :
Dans cette thèse, nous nous intéressons au problème de l'analyse urbaine à partir d'images satellitales par des méthodes automatiques ou semi-automatiques issues du traitement d'image. Dans le premier chapitre, nous présentons le contexte dans lequel le travail a été effectué. Nous exposons les types de données utilisées, les approches statistiques considérées. Nous donnons également quelques exemples d'applications qui justifient une telle étude. Enfin, un état de l'art des diverses méthodes d'analyse des textures est présenté. Dans les deux chapitres suivants, nous développons une méthode automatique d'extraction d'un masque urbain à partir d'une analyse de la texture de l'image. Des méthodes d'extraction d'un masque urbain sont décrites. Ensuite, nous définissons plus précisemment les huit modèles markoviens gaussiens fondés sur des chaines. Ces modèles sont renormalisés par une méthode de renormalisation de groupe issue de la physique statistique afin de corriger le biais introduit par l'anisotropie du réseau de pixels. L'analyse de texture proposée est comparée avec deux méthodes classiques: les matrices de cooccurrence et les filtres de Gabor. L'image du paramètre de texture est ensuite classifiée avec un algorithme non supervisé de classification floue fondée sur la définition d'un critère entropique. Les paramètres estimés avec cet algorithme sont intégrés dans un modèle markovien de segmentation. Des résultats d'extraction de masques urbains sont finalement présentés sur des images satellitales optiques SPOT3, des simulations SPOT5, et des images radar ERS1. Dans le quatrième chapitre, nous présentons l'analyse granulométrique utilisée pour analyser le paysage urbain. Les outils et définitions de base de la morphologie mathématique sont exposés. Nous nous intéressons plus particulièrement à l'ouverture par reconstruction qui est utilisée comme transformation de base de la granulométrie. L'étape de quantification qui suit tout étape de transformation nous permet d'estimer en chaque pixel une distribution locale de taille qui est intégrée dans le terme d'attache aux données d'un modèle markovien de segmentation. Des tests sont effectués sur des simulations SPOT5. |
Abstract :
In this thesis, we investigate the problem of urban areas analysis from satellite images by automatic or semi-automatic methods coming from image processing. In the first chapter, we describe the context of this work, i.e. the type of used data, the statistical applied methods. We also give some examples of the applications which require such an analysis. Finally, a study of the existing methods of texture analysis is presented. In the second and third chapter, we develop a non supervised method based on texture analysis in order to extract an urban mask. First a study of the existing methods of urban mask extraction is presented. Second we precisely describe the eight chain-based Gaussian Markovian models used to characterize urban texture. These models are normalized through a renormalization group technique derived from statistical physics in order to correct the bias introduced by the anisotropy of the lattice.The above mentionned method of texture analysis is then compared with two classical ones: coocurrences matrix and Gabor filters. The image is then partitionned by an unsupervised fuzzy Cmeans algorithm based on an entropic criterion. The final segmentation is performed by the minimization of an energy derived from a Markovian model. Some results are presented that are obtained from SPOT3 images, SPOT5 simulations and radar ERS1 images. In the fourth chapter, we present the granulometric approach used to segment within the urban area itself. The basic operations and definitions of mathematical morphology are settled. We are particularly interested in opening by reconstruction operation based on geodesic dilatations. In fact this operation is used to define a granulometry. The quantification step that follows the transformation step consists in estimating a local size distribution function for each pixel. These parameters are then integrated in the data term of a Markovian model. Some results on SPOT5 simulations are presented. |
|
top of the page
11 Conference articles |
1 - SAR image classification with non- stationary multinomial logistic mixture of amplitude and texture densities.
K. Kayabol and A. Voisin and J. Zerubia. In Proc. IEEE International Conference on Image Processing (ICIP), pages 173-176, Brussels, Belgium, September 2011.
Keywords : High resolution SAR images, Classification, Texture, Multinomial logistic, Classification EM algorithm.
@INPROCEEDINGS{inria-00592252,
|
| author |
= |
{Kayabol, K. and Voisin, A. and Zerubia, J.}, |
| title |
= |
{SAR image classification with non- stationary multinomial logistic mixture of amplitude and texture densities}, |
| year |
= |
{2011}, |
| month |
= |
{September}, |
| booktitle |
= |
{Proc. IEEE International Conference on Image Processing (ICIP)}, |
| pages |
= |
{173-176}, |
| address |
= |
{Brussels, Belgium}, |
| url |
= |
{http://hal.inria.fr/inria-00592252/en/}, |
| keyword |
= |
{High resolution SAR images, Classification, Texture, Multinomial logistic, Classification EM algorithm} |
| } |
Abstract :
We combine both amplitude and texture statistics of the Synthetic Aperture Radar (SAR) images using Products of Experts (PoE) approach for classification purpose. We use Nakagami density to model the class amplitudes. To model the textures of the classes, we exploit a non-Gaussian Markov Random Field (MRF) texture model with t-distributed regression error. Non-stationary Multinomial Logistic (MnL) latent class label model is used as a mixture density to obtain spatially smooth class segments. We perform the classification Expectation-Maximization (CEM) algorithm to estimate the class parameters and classify the pixels. We obtained some classification results of water, land and urban areas in both supervised and semi-supervised cases on TerraSAR-X data. |
|
2 - Nonlinear models for the statistics of adaptive wavelet packet coefficients of texture.
J. Aubray and I. H. Jermyn and J. Zerubia. In Proc. European Signal Processing Conference (EUSIPCO), Florence, Italy, September 2006.
Keywords : Texture, Adaptive, Wavelet packet, Nonlinear, Bimodal, Statistics.
@INPROCEEDINGS{aubray_eusipco06,
|
| author |
= |
{Aubray, J. and Jermyn, I. H. and Zerubia, J.}, |
| title |
= |
{Nonlinear models for the statistics of adaptive wavelet packet coefficients of texture}, |
| year |
= |
{2006}, |
| month |
= |
{September}, |
| booktitle |
= |
{Proc. European Signal Processing Conference (EUSIPCO)}, |
| address |
= |
{Florence, Italy}, |
| pdf |
= |
{ftp://ftp-sop.inria.fr/ariana/Articles/2006_aubray_eusipco06.pdf}, |
| keyword |
= |
{Texture, Adaptive, Wavelet packet, Nonlinear, Bimodal, Statistics} |
| } |
Abstract :
Probabilistic adaptive wavelet packet models of
texture pro- vide new insight into texture structure
and statistics by focus- ing the analysis on
significant structure in frequency space. In very
adapted subbands, they have revealed new bimodal
statistics, corresponding to the structure inherent to
a texture, and strong dependencies between such
bimodal sub- bands, related to phase coherence in a
texture. Existing models can capture the former but
not the latter. As a first step to- wards modelling
the joint statistics, and in order to simplify earlier
approaches, we introduce a new parametric family of
models capable of modelling both bimodal and unimodal
subbands, and of being generalized to capture the
joint statistics. We show how to compute MAP estimates
for the adaptive basis and model parameters, and apply
the models to Brodatz textures to illustrate their
performance. |
|
3 - Texture-adaptive mother wavelet selection for texture analysis.
G.C.K. Abhayaratne and I. H. Jermyn and J. Zerubia. In Proc. IEEE International Conference on Image Processing (ICIP), Genoa, Italy, September 2005.
Keywords : Texture, Wavelet packet, Adaptive, Mother.
@INPROCEEDINGS{abhayaratne_icip05,
|
| author |
= |
{Abhayaratne, G.C.K. and Jermyn, I. H. and Zerubia, J.}, |
| title |
= |
{Texture-adaptive mother wavelet selection for texture analysis}, |
| year |
= |
{2005}, |
| month |
= |
{September}, |
| booktitle |
= |
{Proc. IEEE International Conference on Image Processing (ICIP)}, |
| address |
= |
{Genoa, Italy}, |
| pdf |
= |
{http://www-sop.inria.fr/members/Ian.Jermyn/publications/Abhayaratne05icip.pdf}, |
| keyword |
= |
{Texture, Wavelet packet, Adaptive, Mother} |
| } |
Abstract :
Classification results obtained using wavelet-based texture analysis techniques vary with the choice of mother wavelet used in the methodology. We discuss the use of mother wavelet filters as parameters in a probabilistic approach to texture analysis based on adaptive biorthogonal wavelet packet bases. The optimal choice for the mother wavelet filters is estimated from the data, in addition to the other model parameters. The model is applied to the classification of single texture images and mosaics of Brodatz textures, the results showing improvement over the performance of standard wavelets for a given filter length. |
|
4 - Multimodal statistics of adaptive wavelet packet coefficients: experimental evidence and theory.
R. Cossu and I. H. Jermyn and J. Zerubia. In Proc. Physics in Signal and Image Processing, Toulouse, France, January 2005.
Keywords : Bimodal, Statistics, Wavelet packet, Adaptive, Texture.
@INPROCEEDINGS{cossu_psip05,
|
| author |
= |
{Cossu, R. and Jermyn, I. H. and Zerubia, J.}, |
| title |
= |
{Multimodal statistics of adaptive wavelet packet coefficients: experimental evidence and theory}, |
| year |
= |
{2005}, |
| month |
= |
{January}, |
| booktitle |
= |
{Proc. Physics in Signal and Image Processing}, |
| address |
= |
{Toulouse, France}, |
| pdf |
= |
{http://www-sop.inria.fr/members/Ian.Jermyn/publications/Cossu05psip.pdf}, |
| keyword |
= |
{Bimodal, Statistics, Wavelet packet, Adaptive, Texture} |
| } |
Abstract :
In recent work, it was noted that although the subband histograms
for standard wavelet coefcients take on a generalized
Gaussian form, this is no longer true for wavelet packet
bases adapted to a given texture. Instead, three types of subband
statistics are observed: Gaussian, generalized Gaussian,
and interestingly, in some subbands, bi- or multi-modal histograms.
Motivated by this observation, we provide additional
experimental conrmation of the existence of multimodal
subbands, and provide a theoretical explanation for
their occurrence. The results reveal the connection of such
subbands with the characteristic structure in a texture, and
thus confirm the importance of such subbands for image modelling
and applications. |
|
5 - Texture discrimination using multimodal wavelet packet subbands.
R. Cossu and I. H. Jermyn and J. Zerubia. In Proc. IEEE International Conference on Image Processing (ICIP), Singapore, October 2004.
Keywords : Bimodal, Adaptive, probabilistic, Wavelet packet, Texture.
@INPROCEEDINGS{cossu_icip04,
|
| author |
= |
{Cossu, R. and Jermyn, I. H. and Zerubia, J.}, |
| title |
= |
{Texture discrimination using multimodal wavelet packet subbands}, |
| year |
= |
{2004}, |
| month |
= |
{October}, |
| booktitle |
= |
{Proc. IEEE International Conference on Image Processing (ICIP)}, |
| address |
= |
{Singapore}, |
| pdf |
= |
{http://www-sop.inria.fr/members/Ian.Jermyn/publications/Cossu04icip.pdf}, |
| keyword |
= |
{Bimodal, Adaptive, probabilistic, Wavelet packet, Texture} |
| } |
Abstract :
The subband histograms of wavelet packet bases adapted to individual
texture classes often fail to display the leptokurtotic behaviour
shown by the standard wavelet coefcients of `natural'
images. While many subband histograms remain leptokurtotic
in adaptive bases, some subbands are Gaussian. Most interestingly,
however, some subbands show multimodal behaviour, with
no mode at zero. In this paper, we provide evidence for the existence
of these multimodal subbands and show that they correspond
to narrow frequency bands running throughout images of the texture.
They are thus closely linked to the texture's structure. As
such, they seem likely to possess superior descriptive and discriminative
power as compared to unimodal subbands. We demonstrate
this using both Brodatz and remote sensing images. |
|
6 - Texture analysis using adaptative biorthogonal wavelet packets.
G.C.K. Abhayaratne and I. H. Jermyn and J. Zerubia. In Proc. IEEE International Conference on Image Processing (ICIP), Singapore, October 2004.
Keywords : Adaptive, Wavelet packet, Biorthogonal, Texture, Statistics.
@INPROCEEDINGS{Abhayratne_icip04,
|
| author |
= |
{Abhayaratne, G.C.K. and Jermyn, I. H. and Zerubia, J.}, |
| title |
= |
{Texture analysis using adaptative biorthogonal wavelet packets}, |
| year |
= |
{2004}, |
| month |
= |
{October}, |
| booktitle |
= |
{Proc. IEEE International Conference on Image Processing (ICIP)}, |
| address |
= |
{Singapore}, |
| pdf |
= |
{http://www-sop.inria.fr/members/Ian.Jermyn/publications/Abhayaratne04icip.pdf}, |
| keyword |
= |
{Adaptive, Wavelet packet, Biorthogonal, Texture, Statistics} |
| } |
Abstract :
We discuss the use of adaptive biorthogonal wavelet packet bases
in a probabilistic approach to texture analysis, thus combining the
advantages of biorthogonal wavelets (FIR, linear phase) with those
of a coherent texture model. The computation of the probability
uses both the primal and dual coefcients of the adapted biorthogonal
wavelet packet basis. The computation of the biorthogonal
wavelet packet coefcients is done using a lifting scheme, which
is very efficient. The model is applied to the classification of mosaics
of Brodatz textures, the results showing improvement over
the performance of the corresponding orthogonal wavelets. |
|
7 - Texture analysis using probabilistic models of the unimodal and multimodal statistics of adaptative wavelet packet coefficients.
R. Cossu and I. H. Jermyn and J. Zerubia. In Proc. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Montreal, Canada, May 2004.
Keywords : Bimodal, Adaptive, Wavelet packet, Texture, Gaussian mixture, Statistics.
@INPROCEEDINGS{cossu04a,
|
| author |
= |
{Cossu, R. and Jermyn, I. H. and Zerubia, J.}, |
| title |
= |
{Texture analysis using probabilistic models of the unimodal and multimodal statistics of adaptative wavelet packet coefficients}, |
| year |
= |
{2004}, |
| month |
= |
{May}, |
| booktitle |
= |
{Proc. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, |
| address |
= |
{Montreal, Canada}, |
| pdf |
= |
{http://www-sop.inria.fr/members/Ian.Jermyn/publications/Cossu04icassp.pdf}, |
| keyword |
= |
{Bimodal, Adaptive, Wavelet packet, Texture, Gaussian mixture, Statistics} |
| } |
Abstract :
Although subband histograms of the wavelet coefficients of
natural images possess a characteristic leptokurtotic form,
this is no longer true for wavelet packet bases adapted to
a given texture. Instead, three types of subband statistics
are observed: Gaussian, leptokurtotic, and interestingly, in
some subbands, multimodal histograms. These subbands
are closely linked to the structure of the texture, and guarantee
that the most probable image is not flat. Motivated by
these observations, we propose a probabilistic model that
takes them into account. Adaptive wavelet packet subbands
are modelled as Gaussian, generalized Gaussian, or a constrained
Gaussian mixture. We use a Bayesian methodology,
finding MAP estimates for the adaptive basis, for subband
model selection, and for subband model parameters.
Results confirm the effectiveness of the proposed approach,
and highlight the importance of multimodal subbands for
texture discrimination and modelling. |
|
8 - Texture Analysis: An Adaptive Probabilistic Approach.
K. Brady and I. H. Jermyn and J. Zerubia. In Proc. IEEE International Conference on Image Processing (ICIP), Barcelona, Spain, September 2003.
Keywords : Adaptive, Wavelet packet, Statistics, Texture.
@INPROCEEDINGS{Brady03,
|
| author |
= |
{Brady, K. and Jermyn, I. H. and Zerubia, J.}, |
| title |
= |
{Texture Analysis: An Adaptive Probabilistic Approach}, |
| year |
= |
{2003}, |
| month |
= |
{September}, |
| booktitle |
= |
{Proc. IEEE International Conference on Image Processing (ICIP)}, |
| address |
= |
{Barcelona, Spain}, |
| pdf |
= |
{http://www-sop.inria.fr/members/Ian.Jermyn/publications/Brady03icip.pdf}, |
| keyword |
= |
{Adaptive, Wavelet packet, Statistics, Texture} |
| } |
Abstract :
Two main issues arise when working in the area of texture
segmentation: the need to describe the texture accurately by
capturing its underlying structure, and the need to perform
analyses on the boundaries of textures. Herein, we tackle
these problems within a consistent probabilistic framework.
Starting from a probability distribution on the space of infinite
images, we generate a distribution on arbitrary finite
regions by marginalization. For a Gaussian distribution, the
computational requirement of diagonalization and the modelling
requirement of adaptivity together lead naturally to
adaptive wavelet packet models that capture the ‘significant
amplitude features’ in the Fourier domain. Undecimated
versions of the wavelet packet transform are used to diagonalize
the Gaussian distribution efficiently, albeit approximately.
We describe the implementation and application of
this approach and present results obtained on several Brodatz
texture mosaics. |
|
9 - Adaptive Probabilistic Models of Wavelet Packets for the Analysis and Segmentation of Textured Remote Sensing Images.
K. Brady and I. H. Jermyn and J. Zerubia. In Proc. British Machine Vision Conference (BMVC), Norwich, U. K., September 2003.
Keywords : probabilistic, Adaptive, wavelet, Texture.
@INPROCEEDINGS{Brady03a,
|
| author |
= |
{Brady, K. and Jermyn, I. H. and Zerubia, J.}, |
| title |
= |
{Adaptive Probabilistic Models of Wavelet Packets for the Analysis and Segmentation of Textured Remote Sensing Images}, |
| year |
= |
{2003}, |
| month |
= |
{September}, |
| booktitle |
= |
{Proc. British Machine Vision Conference (BMVC)}, |
| address |
= |
{Norwich, U. K.}, |
| pdf |
= |
{http://www-sop.inria.fr/members/Ian.Jermyn/publications/Brady03bmvc.pdf}, |
| keyword |
= |
{probabilistic, Adaptive, wavelet, Texture} |
| } |
Abstract :
Remote sensing imagery plays an important role in many elds. It has
become an invaluable tool for diverse applications ranging from cartography
to ecosystem management. In many of the images processed in these types
of applications, semantic entities in the scene are correlated with textures
in the image. In this paper, we propose a new method of analysing such
textures based on adaptive probabilistic models of wavelet packets. Our approach
adapts to the principal periodicities present in the textures, and can
capture long-range correlations while preserving the independence of the
wavelet packet coefcients. This technique has been applied to several remote
sensing images, the results of which are presented. |
|
10 - Gaussian Mixture Models of Texture and Colour for Image Database Retrieval.
H. Permuter and J.M. Francos and I. H. Jermyn. In Proc. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Hong Kong, April 2003.
Keywords : Texture, Gaussian mixture, Classification, Aerial images.
@INPROCEEDINGS{Permuter03,
|
| author |
= |
{Permuter, H. and Francos, J.M. and Jermyn, I. H.}, |
| title |
= |
{Gaussian Mixture Models of Texture and Colour for Image Database Retrieval}, |
| year |
= |
{2003}, |
| month |
= |
{April}, |
| booktitle |
= |
{Proc. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, |
| address |
= |
{Hong Kong}, |
| pdf |
= |
{http://www-sop.inria.fr/members/Ian.Jermyn/publications/Permuter03icassp.pdf}, |
| keyword |
= |
{Texture, Gaussian mixture, Classification, Aerial images} |
| } |
Abstract :
We introduce Gaussian mixture models of ‘structure’ and
colour features in order to classify coloured textures in images,
with a view to the retrieval of textured colour images
from databases. Classifications are performed separately
using structure and colour and then combined using
a confidence criterion. We apply the models to the VisTex
database and to the classification of man-made and natural
areas in aerial images. We compare these models with others
in the literature, and show an overall improvement in
performance. |
|
11 - Unsupervised Image Segmentation via Markov Trees and Complex Wavelets.
C. Shaffrey and N. Kingsbury and I. H. Jermyn. In Proc. IEEE International Conference on Image Processing (ICIP), Rochester, USA, September 2002.
Keywords : Segmentation, Hidden Markov Model, Texture, Colour.
@INPROCEEDINGS{ijking,
|
| author |
= |
{Shaffrey, C. and Kingsbury, N. and Jermyn, I. H.}, |
| title |
= |
{Unsupervised Image Segmentation via Markov Trees and Complex Wavelets}, |
| year |
= |
{2002}, |
| month |
= |
{September}, |
| booktitle |
= |
{Proc. IEEE International Conference on Image Processing (ICIP)}, |
| address |
= |
{Rochester, USA}, |
| pdf |
= |
{http://www-sop.inria.fr/members/Ian.Jermyn/publications/Shaffrey02icip.pdf}, |
| keyword |
= |
{Segmentation, Hidden Markov Model, Texture, Colour} |
| } |
Abstract :
The goal in image segmentation is to label pixels in an image based
on the properties of each pixel and its surrounding region. Recently
Content-Based Image Retrieval (CBIR) has emerged as an
application area in which retrieval is attempted by trying to gain
unsupervised access to the image semantics directly rather than
via manual annotation. To this end, we present an unsupervised
segmentation technique in which colour and texture models are
learned from the image prior to segmentation, and whose output
(including the models) may subsequently be used as a content
descriptor in a CBIR system. These models are obtained in a
multiresolution setting in which Hidden Markov Trees (HMT) are
used to model the key statistical properties exhibited by complex
wavelet and scaling function coefficients. The unsupervised Mean
Shift Iteration (MSI) procedure is used to determine a number of
image regions which are then used to train the models for each
segmentation class. |
|
top of the page
13 Technical and Research Reports |
1 - Unsupervised amplitude and texture based classification of SAR images with multinomial latent model.
K. Kayabol and J. Zerubia. Research Report 7700, INRIA, July 2011.
Keywords : High resolution SAR, Classification, Texture.
@TECHREPORT{Kayabol11,
|
| author |
= |
{Kayabol, K. and Zerubia, J.}, |
| title |
= |
{Unsupervised amplitude and texture based classification of SAR images with multinomial latent model}, |
| year |
= |
{2011}, |
| month |
= |
{July}, |
| institution |
= |
{INRIA}, |
| type |
= |
{Research Report}, |
| number |
= |
{7700}, |
| url |
= |
{http://hal.archives-ouvertes.fr/hal-00612491/fr/}, |
| keyword |
= |
{High resolution SAR, Classification, Texture} |
| } |
Abstract :
We combine both amplitude and texture statistics of the Synthetic Aperture Radar (SAR) images using Products of Experts (PoE) approach for classification purpose. We use Nakagami density to model the class amplitudes and a non-Gaussian Markov Random Field (MRF) texture model with t-distributed regression error to model the textures of the classes. A non-stationary Multinomial Logistic (MnL) latent class label model is used as a mixture density to obtain spatially smooth class segments. The Classification Expectation-Maximization (CEM) algorithm is performed to estimate the class parameters and to classify the pixels. We resort to Integrated Classification Likelihood (ICL) criterion to determine the number of classes in the model. We obtained some classification results of water, land and urban areas in both supervised and unsupervised cases on TerraSAR-X, as well as COSMO-SkyMed data.
|
|
2 - Hierarchical finite-state modeling for texture segmentation with application to forest classification.
G. Scarpa and M. Haindl and J. Zerubia. Research Report 6066, INRIA, INRIA, France, December 2006.
Keywords : Texture, Segmentation, Co-occurrence matrix, Structural approach, MCMC, Synthesis.
@TECHREPORT{scarparr06,
|
| author |
= |
{Scarpa, G. and Haindl, M. and Zerubia, J.}, |
| title |
= |
{Hierarchical finite-state modeling for texture segmentation with application to forest classification}, |
| year |
= |
{2006}, |
| month |
= |
{December}, |
| institution |
= |
{INRIA}, |
| type |
= |
{Research Report}, |
| number |
= |
{6066}, |
| address |
= |
{INRIA, France}, |
| url |
= |
{https://hal.inria.fr/inria-00118420}, |
| keyword |
= |
{Texture, Segmentation, Co-occurrence matrix, Structural approach, MCMC, Synthesis} |
| } |
Abstract :
In this research report we present a new model for texture representation which is particularly well suited for image analysis and segmentation. Any image is first discretized and then a hierarchical finite-state region-based model is automatically coupled with the data by means of a sequential optimization scheme, namely the Texture Fragmentation and Reconstruction (TFR) algorithm. The TFR algorithm allows to model both intra- and inter-texture interactions, and eventually addresses the segmentation task in a completely unsupervised manner. Moreover, it provides a hierarchical output, as the user may decide the scale at which the segmentation has to be given. Tests were carried out on both natural texture mosaics provided by the Prague Texture Segmentation Datagenerator Benchmark and remote-sensing data of forest areas provided by the French National Forest Inventory (IFN). |
|
3 - Models of the Unimodal and Multimodal Statistics of Adaptive Wavelet Packet Coefficients.
R. Cossu and I. H. Jermyn and K. Brady and J. Zerubia. Research Report 5122, INRIA, France, February 2004.
Keywords : Wavelet packet, Texture.
@TECHREPORT{5122,
|
| author |
= |
{Cossu, R. and Jermyn, I. H. and Brady, K. and Zerubia, J.}, |
| title |
= |
{Models of the Unimodal and Multimodal Statistics of Adaptive Wavelet Packet Coefficients}, |
| year |
= |
{2004}, |
| month |
= |
{February}, |
| institution |
= |
{INRIA}, |
| type |
= |
{Research Report}, |
| number |
= |
{5122}, |
| address |
= |
{France}, |
| url |
= |
{https://hal.inria.fr/inria-00071461}, |
| pdf |
= |
{https://hal.inria.fr/file/index/docid/71461/filename/RR-5122.pdf}, |
| ps |
= |
{https://hal.inria.fr/docs/00/07/14/61/PS/RR-5122.ps}, |
| keyword |
= |
{Wavelet packet, Texture} |
| } |
Résumé :
De récents travaux ont montré que bien que les histogrammes de sous-bandes pour les coefficients d'ondelettes standards ont une forme de gaussienne généralisée, ce n'est plus vrai pour les bases de paquets d'ondelettes adaptés à une certaine texture. Trois types de statistiques sont alors observés pour les sous-bandes: gaussienne, gaussienne generalisée et dans certaines sous-bandes des histogrammes multimodaux sans mode en zéro. Dans ce rapport, nous démontrons que ces sous-bandes sont étroitement liées à la structure de la texture et sont ainsi primordiales dans les applications dans lesquelles la texture joue un rôle important. Fort de ces observations, nous étendons l'approche de modélisation de textures proposée par en incluant ces sous-bandes. Nous modifions l'hypothèse gaussienne pour inclure les gaussiennes généralisées et les mixtures de gaussiennes contraintes. Nous utilisons une méthodologie bayésienne, définissant des estimateurs MAP pour la base adaptative, pour la sélection du modèle de la sous-bande et pour les paramètres de ce modèle. Les résultats confirment l'efficacité de la méthode proposée et soulignent l'importance des sous-bandes multimodales pour la discrimination et la modélisation de textures. |
Abstract :
In recent work, it was noted that although the subband histograms for standard wavelet coefficients take on a generalized Gaussian form, this is no longer true for wavelet packet bases adapted to a given texture. Instead, three types of subband statistics are observed: Gaussian, generalized Gaussian, and most interestingly, in some subbands, multimodal histograms with no mode at zero. As will be demonstrated in this report, these latter subbands are closely linked to the structure of the texture, and are thus likely to be important for many applications in which texture plays a role. Motivated by these observations, we extend the approach to texture modelling proposed by to include these subbands. We relax the Gaussian assumption to include generalized Gaussians and constrained Gaussian mixtures. We use a Bayesian methodology, finding MAP estimates for the adaptive basis, for subband model selection, and for subband model parameters. Results confirm the effectiveness of the proposed approach, and highlight the importance of multimodal subbands for texture discrimination and modelling. |
|
4 - Structure and Texture Compression.
J.F. Aujol and B. Matei. Research Report 5076, INRIA, France, January 2004.
Keywords : Bounded Variation Space, Image decomposition, Texture, Structure.
@TECHREPORT{5076,
|
| author |
= |
{Aujol, J.F. and Matei, B.}, |
| title |
= |
{Structure and Texture Compression}, |
| year |
= |
{2004}, |
| month |
= |
{January}, |
| institution |
= |
{INRIA}, |
| type |
= |
{Research Report}, |
| number |
= |
{5076}, |
| address |
= |
{France}, |
| url |
= |
{https://hal.inria.fr/inria-00071507}, |
| pdf |
= |
{https://hal.inria.fr/file/index/docid/71507/filename/RR-5076.pdf}, |
| ps |
= |
{https://hal.inria.fr/docs/00/07/15/07/PS/RR-5076.ps}, |
| keyword |
= |
{Bounded Variation Space, Image decomposition, Texture, Structure} |
| } |
Résumé :
Dans ce papier, nous nous intéressons au problème de la compression d'image. Les ondelettes se sont révélées être un outil particulièremment efficace . Récemment, de nombreux algorithmes ont été proposés pour amméliorer la compression par ondelettes en essayant de prendre en compte les strucutres présentes dans l'image. De telles méthodes se révèlents très efficaces pour les images géométriques. Nous construisons un algorithme de compression d'images qui prend en compte la géométrie de l'image tout en étant capable d'être performant sur des images contenant à la fois des structures et des textures. Pour cela, nous utilisons un algorithme de décomposition d'image récemment introduit dans . Cet algorithme permet de séparer une image en deux composantes, une première composante contenant l'information géométrique de l'image, et une deuxième contenant les éléments oscillants de l'image. L'idée de notre méthode de compression est la suivante. Nous commen ons par décomposer l'image à compresser en sa partie géométrique et sa partie oscillante. Nous effectuons ensuite la compression de la partie géométrique à l'aide de l'algorithme introduit dans , ce dernier étant particulièrement bien adapté pour la compression des structures d'une image. Pour la partie oscillante de l'image, nous utilisons l'algorithme classique de compression par ondelettes biorthogonales. sur les zones régulières d'une image). l'image. Notre nouvel algorithme de compression s'avère plus performant que la méthode classique par ondelettes biorthogonales. meilleurs à la fois en PSNR, et aussi visuellement (les bords sont plus précis et les textures sont mieux conservées). |
Abstract :
In this paper, we tackle the problem of image compression. During the last past years, many algorithms have been proposed to take advantage of the geometry of the image. We intend here to propose a new compression algorithm which would take into account the structures in the image, and which would be powerful even when the original image has some textured areas. To this end, we first split our image into two components, a first one containing the structures of the image, and a second one the oscillating patterns. We then perform the compression of each component separately. Our final compressed image is the sum of these two compressed components. This new compression algorithm outperforms the standard biorthogonal wavelets compession. |
|
5 - A Probabilistic Framework for Adaptive Texture Description.
K. Brady and I. H. Jermyn and J. Zerubia. Research Report 4920, INRIA, France, September 2003.
Keywords : Segmentation, Texture, Wavelet packet.
@TECHREPORT{4920,
|
| author |
= |
{Brady, K. and Jermyn, I. H. and Zerubia, J.}, |
| title |
= |
{A Probabilistic Framework for Adaptive Texture Description}, |
| year |
= |
{2003}, |
| month |
= |
{September}, |
| institution |
= |
{INRIA}, |
| type |
= |
{Research Report}, |
| number |
= |
{4920}, |
| address |
= |
{France}, |
| url |
= |
{https://hal.inria.fr/inria-00071659}, |
| pdf |
= |
{https://hal.inria.fr/file/index/docid/71659/filename/RR-4920.pdf}, |
| ps |
= |
{https://hal.inria.fr/docs/00/07/16/59/PS/RR-4920.ps}, |
| keyword |
= |
{Segmentation, Texture, Wavelet packet} |
| } |
Résumé :
Ce rapport présente le développement d'un nouveau cadre probabiliste cohérent pour la description adaptative de texture. En partant d'une distribution de probabilité sur un espace d'images infinies, nous générons une distribution sur des régions finies par marginalisation. Pour une distribution gaussienne, les contraintes de calcul imposées par la diagonalisation nous conduisent naturellement à des modèles utilisant des paquets d'ondelettes adaptatifs. Ces modèles reflètent les principales périodicités présentes dans les textures et permettent également d'avoir des corrélations à longue portée tout en préservant l'indépendance des coefficients des paquets d'ondelettes. Nous avons appliqué notre méthode à la segmentation. Deux types de données figurent dans notre ensemble de test: des mosaïques synthétiques de Brodatz et des images satellitaires haute résolution. Dans le cas des textures synthétiques, nous utilisons la version non-décimée de la transformée en paquets d'ondelettes afin de diagonaliser la distribution gaussienne de manière efficace, bien qu'approximative. Cela nous permet d'effectuer une classification de la mosaique pixel par pixel. Une étape de régularisation est ensuite effectuée afin d'arriver à un résultat de segmentation final plus lisse. Afin d'obtenir les meilleurs résultats possibles dans le cas de données réelles, la moyenne de la distribution est ensuite introduite dans le modèle. L'approximation faite pour la classification des mosaiques de textures synthetiques a été testée sur des images réelles, mais les résultats obtenus n'étaient pas satisfaisants. C'est pourquoi nous avons introduit, pour ce type de données, une technique de classification heuristique basée sur la transformée en paquets d'ondelettes décimée. Les résultats de segmentation sont ensuite régularisés à l'aide de la même méthode que dans le cas synthétique. Nous présentons les résultats pour chaque type de données et concluons par une discussion. |
Abstract :
This report details the development of a probabilistic framework for adaptive texture description. Starting with a probability distribution on the space of infinite images, we generate a distribution on finite regions by marginalisation. For a Gaussian distribution, the computational requirement of diagonalisation leads naturally to adaptive wavelet packet models which capture the principal periodicities present in the textures and allow long-range correlations while preserving the independence of the wavelet packet coefficients. These models are then applied to the task of segmentation. Two data types are included in our test bed: synthetic Brodatz mosaics and high-resolution satellite images. For the case of the synthetic textures, undecimated versions of the wavelet packet transform are used to diagonalise the Gaussian distribution efficiently, albeit approximately. This enables us to perform a pixelwise classification of the mosaics. A regularisation step is then implemented in order to arrive at a smooth final segmentation. In order to obtain the best possible results for the real dataset, the mean of the distribution is included in the model. The approximation made for the classification of the synthetic texture mosaics is tested on the remote sensing images, but it produces unsatisfactory results. Therefore we introduce a heuristic classification technique for this dataset, based on a decimated wavelet packet transform. The resulting segmentation is then regularised using the same method as in the synthetic case. Results are presented for both types of data and a discussion follows. |
|
6 - Image Decomposition : Application to Textured Images and SAR Images.
J.F. Aujol and G. Aubert and L. Blanc-Féraud and A. Chambolle. Research Report 4704, INRIA, France, January 2003.
Keywords : Total variation, Bounded Variation Space, Texture, Classification, Restoration, Synthetic Aperture Radar (SAR).
@TECHREPORT{4704,
|
| author |
= |
{Aujol, J.F. and Aubert, G. and Blanc-Féraud, L. and Chambolle, A.}, |
| title |
= |
{Image Decomposition : Application to Textured Images and SAR Images}, |
| year |
= |
{2003}, |
| month |
= |
{January}, |
| institution |
= |
{INRIA}, |
| type |
= |
{Research Report}, |
| number |
= |
{4704}, |
| address |
= |
{France}, |
| url |
= |
{https://hal.inria.fr/inria-00071882}, |
| pdf |
= |
{https://hal.inria.fr/file/index/docid/71882/filename/RR-4704.pdf}, |
| ps |
= |
{https://hal.inria.fr/docs/00/07/18/82/PS/RR-4704.ps}, |
| keyword |
= |
{Total variation, Bounded Variation Space, Texture, Classification, Restoration, Synthetic Aperture Radar (SAR)} |
| } |
Résumé :
Dans ce rapport, nous présentons un nouvel algorithme pour décomposer une imagef en u+v, u étant à variation bornée, et v contenant les textures et le bruit de l'image originale. Nous introduisons une fonctionnelle adaptée à ce problème. Le minimum de cette fonctionnelle correspond à la décomposition cherchée de l'image. Le calcul de ce minimum se fait par minimisation successive par rapport à chacune des variables, chaque minimisati- on étant réalisée à l'aide d'un algorithme de projection. Nous faisons l'étude théorique de notre modèle, et nous présentons des résultats numériques. D'une part, nous montrons comment la composante v peut être utilisée pour faire de la classification d'images texturées, et d'autre part nous montrons comment la composante u peut être utilisée en restauration d'images SAR. |
Abstract :
In this report, we present a new algorithm to split an image f into a component u belonging to BV and a component v made of textures and noise of the initial image. We introduce a functional adapted to this problem. The minimum of this functional corresponds to the image decomposition we want to get. We compute this minimum by minimizing successively our functional with respect to u and v. We carry out the mathematical study of our algorithm. We present some numerical results. On the one hand, we show how the v component can be used to classify textured images, and on the other hand, we show how the u component can be used in SAR image restoration. |
|
7 - Supervised Classification for Textured Images.
J.F. Aujol and G. Aubert and L. Blanc-Féraud. Research Report 4640, Inria, France, November 2002.
Keywords : Texture, Classification, Wavelets, Partial differential equation, Level sets.
@TECHREPORT{4640,
|
| author |
= |
{Aujol, J.F. and Aubert, G. and Blanc-Féraud, L.}, |
| title |
= |
{Supervised Classification for Textured Images}, |
| year |
= |
{2002}, |
| month |
= |
{November}, |
| institution |
= |
{Inria}, |
| type |
= |
{Research Report}, |
| number |
= |
{4640}, |
| address |
= |
{France}, |
| url |
= |
{https://hal.inria.fr/inria-00071945}, |
| pdf |
= |
{https://hal.inria.fr/file/index/docid/71945/filename/RR-4640.pdf}, |
| ps |
= |
{https://hal.inria.fr/docs/00/07/19/45/PS/RR-4640.ps}, |
| keyword |
= |
{Texture, Classification, Wavelets, Partial differential equation, Level sets} |
| } |
Résumé :
Dans ce rapport, nous présentons un modèle de classification supervisée basé sur une approche variationnelle. Ce modèle s'applique spécifiquement aux images texturées. Nous souhaitons obtenir une partition optimale de l'image constituée de textures séparées par des interfaces régulières. Pour cela, nous représentons les régions définies par les classes ainsi que leurs interfaces par des fonctions d'ensemble de niveaux. Nous définissons une fonctionnelle sur ces ensembles de niveaux dont le minimum est une partition optimale. Cette fonctionnelle comporte en particulier un terme d'attache aux données spécifique aux textures. Nous utilisons une transformée en paquets d'ondelettes pour analyser les textures, ces dernières étant caractérisées par la distribution de leur énergie dans chaque sous-bande de la décompositon. Les équations aux dérivées partielles (EDP) relatives à la minimisation de la fonctionnelle sont couplées et plongées dans un schéma dynamique. En fixant un ensemble de niveaux initial, les différents termes des EDP guident l'évolution des interfaces (ensemble de niveau zéro) vers les frontières de la partion optimale, par le biais de forces externes (régularité de l'interface) et internes (attache aux données et contraintes partition). Nous avons effectué des tests sur des images synthétiques et sur des images réelles. |
Abstract :
In this report, we present a supervised classification model based on a variational approach. This model is specifically devoted to textured images. We want to get an optimal partition of an image which is composed of textures separated by regular interfaces. To reach this goal, we represent the regions defined by the classes as well as their interfaces by level set functions. We define a functional on these level sets whose minimizers define an optimal partition. In particular, this functional owns a data term specific to textures. We use a packet wavelet transform to analyze the textures, these ones being characterized by their energy distribution in each sub-band of the decomposition. The partial differential equations (PDE) related to the minimization of the functional are embeded in a dynamical scheme. Given an initial interface set (zero level set), the different terms of the PDE's govern the motion of interfaces such that, at convergence, we get an optimal partition as defined above. Each interface is guided by external forces (regularity of the interface), and internal ones (data term and partition constraints). We have conducted several experiments on both synthetic and real images. |
|
8 - Analyse Intra-urbaine à partir d'Images Satellitaires par une Approche de Fusion de Données sur la Ville de Mexico.
O. Viveros-Cancino and X. Descombes and J. Zerubia. Research Report 4578, Inria, France, October 2002.
Keywords : Data fusion, Markov Fields, Texture, Urban areas, Confusion matrix.
@TECHREPORT{4578,
|
| author |
= |
{Viveros-Cancino, O. and Descombes, X. and Zerubia, J.}, |
| title |
= |
{Analyse Intra-urbaine à partir d'Images Satellitaires par une Approche de Fusion de Données sur la Ville de Mexico}, |
| year |
= |
{2002}, |
| month |
= |
{October}, |
| institution |
= |
{Inria}, |
| type |
= |
{Research Report}, |
| number |
= |
{4578}, |
| address |
= |
{France}, |
| url |
= |
{https://hal.inria.fr/inria-00072010}, |
| pdf |
= |
{https://hal.inria.fr/file/index/docid/72010/filename/RR-4578.pdf}, |
| ps |
= |
{https://hal.inria.fr/docs/00/07/20/10/PS/RR-4578.ps}, |
| keyword |
= |
{Data fusion, Markov Fields, Texture, Urban areas, Confusion matrix} |
| } |
Résumé :
Ce document présente une analyse intra-urbaine afin d'améliorer la détection des différents tissus urbains avec une application sur la ville de Mexico. La méthode de fission-fusion est proposée ainsi qu'une méthode pour fusionner les classes existantes. Les deux méthodes se composent des étapes suivantes : premièrement, une analyse de texture, nommée étape de fission, est faite pour mieux décrire l'image, ensuite, une classification supervisée, nommée étape de fusion, est faite sur les paramètres issus de l'analyse de texture à partir des valeurs de qualité, notamment la valeur Kappa calculée sur la matrice de confusion. Ces étapes sont réalisées sur des images optiques (SPOT) et radar (ERS) de la ville de Mexico et sont suivies d'un régularisation. |
Abstract :
In this research report we present an intra-urban analysis to improve urban texture extraction. Two methods are proposed : a fission-fusion method and another method which fuses already existing classes. Both methods consist of two steps. The first step, called fission, performs a texture analysis which looks for structures with different parameters. The second step, called fusion, involves a supervised classification using quality parameters, in particular the kappa value which is computed from the confusion matrix. These two steps are carried out on SPOT and radar images of Mexico city. A regularization step is then performed which completes our analysis. |
|
9 - Analyse de Texture Hyperspectrale par Modélisation Markovienne.
G. Rellier and X. Descombes and F. Falzon and J. Zerubia. Research Report 4479, INRIA, France, June 2002.
Keywords : Classification, Markov Fields, Texture, Hyperspectral imaging.
@TECHREPORT{4479,
|
| author |
= |
{Rellier, G. and Descombes, X. and Falzon, F. and Zerubia, J.}, |
| title |
= |
{Analyse de Texture Hyperspectrale par Modélisation Markovienne}, |
| year |
= |
{2002}, |
| month |
= |
{June}, |
| institution |
= |
{INRIA}, |
| type |
= |
{Research Report}, |
| number |
= |
{4479}, |
| address |
= |
{France}, |
| url |
= |
{https://hal.inria.fr/inria-00072109}, |
| pdf |
= |
{https://hal.inria.fr/file/index/docid/72109/filename/RR-4479.pdf}, |
| ps |
= |
{https://hal.inria.fr/docs/00/07/21/09/PS/RR-4479.ps}, |
| keyword |
= |
{Classification, Markov Fields, Texture, Hyperspectral imaging} |
| } |
Résumé :
L'analyse de texture est l'objet de nombreuses recherches dans le domaine de l'imagerie mono et multispectrale. En parallèle, sont apparus ces dernières années de nouveaux instruments spectro-imageurs ayant un grand nombre de canaux (supérieur à 10), fournissant des images appelées hyperspectrales qui sont une représentation du paysage échantillonnée à la fois spatialement et spectralement. Le but de ce travail est de réaliser une analyse de texture qui se déroule conjointement dans ces deux espaces discrets. Pour ce faire, on utilise une modélisation probabiliste vectorielle de la texture via un champ de Markov gaussien. Les paramètres de ce champ permettent la caractérisation de différentes textures présentes dans les images hyperspec- trales. L'application visée dans cette étude étant la classification du tissu urbain, qui est mal caractérisée par la seule radiométrie, on utilise ces paramètres comme de nouvelles bandes afin d'effectuer la classification par le critère du Maximum de Vraisemblance. Les résultats sur des images AVIRIS montrent une nette amélioration de la classification due à l'utilisatio- n de l'information de texture. |
Abstract :
Texture analysis has been widely investigated in monospectral and multispectr- al imagery domain. In the same time, new image sensors with a large number of bands (more than 10) have been designed. They are able to provide images with both fine spectral and spatial sampling, called hyperspectral images. The aim of this work is to perform a joint texture analysis in both discrete spaces. To achieve this goal, we have a probabilistic vectorial texture modeling, with Gauss-Markov Random Field. The MRF parameters allow for the characterisation of different hyperspectral textures. A likely application of this work being the classification of urban areas, which are not well characterized by radiometry alone, we use these parameters as new features is a Maximum Likelihood classification algorithm. The results obtain on AVIRIS hyperspectral images show better classifications when using texture information. |
|
10 - La poursuite de projection pour la classification d'image hyperspectrale texturée.
G. Rellier and X. Descombes and F. Falzon and J. Zerubia. Research Report 4152, Inria, France, March 2001.
Keywords : Classification, Texture, Hyperspectral imaging, Markov Fields.
@TECHREPORT{xd01,
|
| author |
= |
{Rellier, G. and Descombes, X. and Falzon, F. and Zerubia, J.}, |
| title |
= |
{La poursuite de projection pour la classification d'image hyperspectrale texturée}, |
| year |
= |
{2001}, |
| month |
= |
{March}, |
| institution |
= |
{Inria}, |
| type |
= |
{Research Report}, |
| number |
= |
{4152}, |
| address |
= |
{France}, |
| url |
= |
{https://hal.inria.fr/inria-00072472}, |
| pdf |
= |
{https://hal.inria.fr/file/index/docid/72472/filename/RR-4152.pdf}, |
| ps |
= |
{https://hal.inria.fr/docs/00/07/24/72/PS/RR-4152.ps}, |
| keyword |
= |
{Classification, Texture, Hyperspectral imaging, Markov Fields} |
| } |
Résumé :
Dans ce travail, nous considérons le problème de la classification supervisée de texture à partir d'images multi-composante de télédetection, dites hyperspectrales. Ces images, le plus souvent acquises par des instruments spectro-imageurs dont le nombre de canaux est en général supérieur à 10, fournissent ainsi une représentation du paysage échantillonnée à la fois spatialement et spectralement. Le but de ce travail est de réaliser une analyse de texture qui se déroule conjointement dans ces deux espaces discrets. On recherche ainsi à enrichir la représentation "habituelle" de texture fondée sur la prise en compte des variations locales de contraste, par l'adjonction d'une connaissance sur ses variations spectrales. L'applicati- on qui est susceptible de bénéficier directement des résultats de cette étude est la classification du tissu urbain. En effet, la réponse spectrale (radiométrique) des zones urbaines est en général ambiguë du fait de la similitude de réponse spectrale de certains matériaux constitutifs du paysage urbain avec certains éléments naturels tels que l'eau, le sol nu, la végétation. La multiplication des bandes spectrales a pour conséquence de rendre plus complexes les mesures et demande également la prise en considération d'un nombre d'échantillons d'apprentissage très important. Quand le nombre de ces échantillons n'est pas suffisant, il faut passer par une étape de réduction de la dimension de l'espace d'observation. Pour prendre en compte le problème de la dimension et celui de l'analyse de texture conjointement dans le domaine spatial et spectral, on se propose ici de faire coopérer un algorithme de poursuite de projection paramétrique, déjà utilisé pour la réduction d'espace dans un cadre non-contextuel, à un modèle de texture par champ markovien, dit modèle markovien gaussien. |
Abstract :
In this work we develop a supervised texture classification algorithm for application to the class of multi-component images called hyperspectral. These images, usually recorded by spectrometers with a number of bands greater than 10, give both a spatially and spectrally sampled representation of a remote scene. The aim of this work is to perform a joint texture analysis in both discrete spaces. The use of spectral variations in this joint texture analysis scheme enables us to improve on the standard representa- tion of textures which only takes into account the local contrast variations. A likely application of this work is urban area classification. Indeed, the spectral response of urban areas is in general ambiguous because some of its constitutive elements have the same reflectance as natural elements such as water, vegetation or bare soil. The greater number of spectral bands makes the measures more complex and so creates the need for a greater number of training samples. When the number of training samples is not sufficient, a necessary step in the analysis is to reduce the dimension of the observation space. To take into account both the problem of dimensional- ity and the jointly spectral and spatial texture analysis, we propose to use in cooperation a projection pursuit algorithm and a Gauss-Markov random field texture model. |
|
11 - Classification d'images satellitaires hyperspectrales en zone rurale et périurbaine.
O. Pony and X. Descombes and J. Zerubia. Research Report 4008, Inria, September 2000.
Keywords : Hyperspectral imaging, Markov Fields, Simulated Annealing, Gibbs Random Fields, Potts model, Texture.
@TECHREPORT{pony00,
|
| author |
= |
{Pony, O. and Descombes, X. and Zerubia, J.}, |
| title |
= |
{Classification d'images satellitaires hyperspectrales en zone rurale et périurbaine}, |
| year |
= |
{2000}, |
| month |
= |
{September}, |
| institution |
= |
{Inria}, |
| type |
= |
{Research Report}, |
| number |
= |
{4008}, |
| url |
= |
{https://hal.inria.fr/inria-00072636}, |
| pdf |
= |
{https://hal.inria.fr/file/index/docid/72636/filename/RR-4008.pdf}, |
| ps |
= |
{https://hal.inria.fr/docs/00/07/26/36/PS/RR-4008.ps}, |
| keyword |
= |
{Hyperspectral imaging, Markov Fields, Simulated Annealing, Gibbs Random Fields, Potts model, Texture} |
| } |
Résumé :
L'observation satellitaire en zone rurale et périurbaine fournit des images hyperspectrales exploitables en vue de réaliser une cartographie ou une analyse du paysage. Nous avons appliqué une classification par maximum de vraisemblance sur des images de zone agricole. Afin de régulariser la classification, nous considérons la modélisation d'image par champs de Markov, dont l'équivalence avec les champs de Gibbs nous permet d'utiliser plusieurs algorithmes itératifs d'optimisation : l'ICM et le recuit simulé, qui convergent respectivement vers une classification sous-optimale ou optimale pour une certaine énergie. Un modèle d'énergie est proposé : le modèle de Potts, que nous améliorons pour le rendre adaptatif aux classes présentes dans l'image. L'étude de la texture dans l'image initiale permet d'introduire des critères artificiels qui s'ajoutent à la radiométrie de l'image en vue d'améliorer la classification. Ceci permet de bien segmenter les zones périurbaines, la forêt, la campagne, dans le cadre d'un plan d'occupation des sols. Trois images hyperspectrales et une vérité terrain ont été utilisées pour réaliser des tests, afin de mettre en évidence les méthodes et le paramétrage adéquats pour obtenir les résultats les plus satisfaisants. |
Abstract :
Satellite observation in rural and semiurban areas provides hyperspectral images which enable us to make a map or an analysis of the landscape. Herein, we applied a maximum likelihood classification on agricultural images. In order to improve this procedure, it is possible in each pixel to use contextual information. Thus, we consider Markov random fields image modeling. The equivalence between Markov and Gibbs fields allows us to use some iterative algorithms of optimisation : ICM and simulated annealing, which converge respectively towards a suboptimal or an optimal classification for a given energy. An energy model is proposed : the Potts model, which can be improved to be adaptive to the classes defined in the image. Texture analysis on the initial image is used to introduce artificial criteria, added to the original image, in order to improve classification. This proves to be useful for segmenting semiurban regions, forests, and the countryside, within the framework of a land-use plan. We use three hyperspectral images and a ground truth to carry out tests, in order to highlight the best methods and parameter setting to obtain the most satisfactory results. |
|
12 - Indexing and retrieval in multimedia libraries through parametric texture modeling using the 2D Wold decomposition.
R. Stoica and J. Zerubia and J.M. Francos. Research Report 3594, Inria, December 1998.
Keywords : Markov Fields, Texture, Segmentation, Indexation.
@TECHREPORT{stoica98,
|
| author |
= |
{Stoica, R. and Zerubia, J. and Francos, J.M.}, |
| title |
= |
{Indexing and retrieval in multimedia libraries through parametric texture modeling using the 2D Wold decomposition}, |
| year |
= |
{1998}, |
| month |
= |
{December}, |
| institution |
= |
{Inria}, |
| type |
= |
{Research Report}, |
| number |
= |
{3594}, |
| url |
= |
{https://hal.inria.fr/inria-00073085}, |
| pdf |
= |
{https://hal.inria.fr/file/index/docid/73085/filename/RR-3594.pdf}, |
| ps |
= |
{https://hal.inria.fr/docs/00/07/30/85/PS/RR-3594.ps}, |
| keyword |
= |
{Markov Fields, Texture, Segmentation, Indexation} |
| } |
Résumé :
Ce rapport présente une méthode paramétrique permettant de faire de l'indexati- on et de la recherche dans une base de données multimédia. L'indexation (étiquetage) et la recherche de données multimédia sont réalisées grâce à la modélisation paramétrique de textures qui se trouvent dans les images de la base de données. Les textures sont caracterisées par des paramètres qui servent d'indices pour la recherche dans la base de données. Afin de pouvoir identifier les différentes régions texturées d'une image et estimer les paramètres correspondants, un algorithme de segmentation-estimatio- n est proposé dans ce rapport, qui fait appel à une décomposition de Wold 2D pour le modèle de texture et à un modèle markovien pour l'étiquetage. L'indexation nécessite de définir une distance entre les images. Une nouvelle distance, inspirée de la distance de Kullback, est décrite dans ce rapport. Elle utilise les paramètres estimés correspondants au modèle 2D de chaque texture. Les résultats obtenus relativement à la segmentation et à l'indexatio- n sont proches de ceux obtenus par un opérateur humain. |
Abstract :
This paper presents a parametric method for indexing and retrieval of multimedia data in digital libraries. %Indexing (labeling) and retrieval %of multimedia data, based on the properties %of the imagery components of the stored data record, are derived. Indexing (labeling) and retrieval of the multimedia data are performed using parametric modeling of the textured segments found in the data imagery components. The estimated parametric models of the textured segments serve as their indices, and hence as indices of the entire image, as well as of the multimedia record which the image is part thereof. To achieve the ability to identify textured image regions and estimate their parameters, a joint segmentation-estimation algorithm that combines the 2-D Wold decomposition based texture model with a Markovian labeling process, is derived. Ordering and indexing of images require a definition of a distance measure between images. Using the framework of the Kullback distance between probability distributions, a new rigorous distance measure between textures is derived. The distance between any two textured image segments is evaluated using their estimated parametric models. The proposed segmentation, distance evaluation, and indexing methods are shown to produce comparable results to those obtained by a human viewer. |
|
13 - Extraction des zones urbaines fondée sur une analyse de la texture par modélisation markovienne.
A. Lorette and X. Descombes and J. Zerubia. Research Report 3423, Inria, May 1998.
Keywords : Texture, Markov Fields, Urban areas, Entropy.
@TECHREPORT{loretteRR98,
|
| author |
= |
{Lorette, A. and Descombes, X. and Zerubia, J.}, |
| title |
= |
{Extraction des zones urbaines fondée sur une analyse de la texture par modélisation markovienne}, |
| year |
= |
{1998}, |
| month |
= |
{May}, |
| institution |
= |
{Inria}, |
| type |
= |
{Research Report}, |
| number |
= |
{3423}, |
| url |
= |
{http://hal.inria.fr/inria-00073267}, |
| pdf |
= |
{http://hal.inria.fr/docs/00/07/32/67/PDF/RR-3423.pdf}, |
| ps |
= |
{http://hal.inria.fr/docs/00/07/32/67/PS/RR-3423.ps}, |
| keyword |
= |
{Texture, Markov Fields, Urban areas, Entropy} |
| } |
Résumé :
Pour délimiter un masque urbain précis à partir d'une image satellitaire la seule information du niveau de gris est insuffisante. Laplupart des méthodes font donc appel à une analyse de la texture de l'image. Nous nous sommes placés dans ce cadre. Dans une première étape, nous avons défini un nouveau paramètre de texture à partir d'un modèle markovien gaussien. Nous obtenons ce nouveau paramètre en calculant la variance conditionnelle de l'image dans huit directions. Ainsi, nous éliminons la mauvaise classification d'objets ayant une orientation privilégiée tels que les vignes et les serres par exemple. Dans une seconde étape, nous proposons un algorithme de emphfuzzy Cmeans modifié incluant un terme d'entropie et pour lequel le nombre de classes n'est pas fixé a priori. Cet algorithme nous permet d'obtenir une première classification de l'image. Enfin, nous régularisons l'image ainsi obtenue grâce à une modélisation par champs de Markov. Des résultats obtenus sur des simulations d'images SPOT5 fournies par le CNES sont présentés. |
Abstract :
Urban areas cannot be extracted from satellite images through only grey level information. Hence most methods analyze the texture of the image to discriminate between urban areas and non urban areas. We define a new texture parameter derived from a Markovian Gaussian model. This new parameter takes into account the variance of the image in eight directions- . Consequently it copes with the misclassification of objects with a privileged orientation like vineyards or greenhouses for instance. Afterwards we develop a modified fuzzy Cmeans algorithm including an entropy term. The advantage of such an algorithm is that the number of classes does not need to be known a priori. By applying this modified fuzzy Cmeans algorithm on the parameter image we obtain a first classification. Finally we regularize the segmented image by using a Markov random field modelling. Some results on SPOT5 simulated images are presented. These images are provided by the CNES (French Space Agency). |
|
top of the page
These pages were generated by 
|
New demo webpage
get the bibtex
