Abstracts


P.-M. Jodoin, M. Mignotte.
Optical-flow based on an edge-avoidance procedure
Computer Vision and Image Understanding,

Abstract: his paper presents a differential optical flow method which accounts for two typical motion-estimation problems : (1) flow regularization within regions of uniform motion while (2) preserving sharp edges near motion discontinuities i.e., where motion is multimodal by nature. The method proposed is a modified version of the well known Lucas Kanade (LK) algorithm. While many edge-preserving strategies try to minimize the effect of outliers by using a line process or a robust function, our method takes a novel approach to solve the problem. Based on documented assumptions, our method computes motion with a classical least-squares fit on a local neighborhood shifted away from where motion is likely to be multimodal. In this way, the inherent bias due to multiple motion around moving edges is avoided instead of being compensated. This edge-avoidance procedure is based on the non-parametric mean-shift algorithm which shifts the LK integration window away from local sharp edges. Our method also locally regularizes motion by performing a fusion of local motion estimates. The regularization is made with a covariance filter which minimizes the effect of uncertainties due in part to noise and/or lack of texture. Our method is compared with other edge-preserving methods on image sequences representing different challenges.

Keywords: Optical flow, motion estimation, information fusion, mean-shift





M. Mignotte.
A non-local regularization strategy for image deconvolution.
Pattern Recognition Letters, 2008.

Abstract: In this paper, we propose an inhomogeneous restoration (deconvolution) model under the Bayesian framework exploiting a non parametric adaptive prior distribution derived from the appealing and natural image model recently proposed by Buades et al. for pure denoising applications. This prior expresses that acceptable restored solutions are likely the images exhibiting a high degree of redundancy. In other words, this prior will favor solutions (i.e., restored images) with similar pixel neighborhood configurations. In order to render}this restoration unsupervised, we have adapted the L-curve approach (originally defined for Tikhonov-type regularizations), for estimating our regularization parameter. The experiments herein reported illustrate the potential of this approach and demonstrate that this regularized restoration strategy performs competitively compared to the best existing state-of-the art methods employing classical local priors (or regularization terms) in benchmark tests.

Keywords: Image deconvolution or restoration, non-local regularization, penalized likelihood, L-curve estimation.





M. Mignotte.
Segmentation by fusion of histogram-based K-means clusters in different color spaces.
IEEE Transactions on Image Processing, 17(5):780-787, May 2008.

Abstract: This paper presents a new, simple, and efficient segmentation approach, based on a fusion procedure which aims at combining several segmentation maps associated to simpler partition models in order to finally get a more reliable and accurate segmentation result. The different label fields to be fused in our application are given by the same and simple ( K-means based) clustering technique on an input image expressed in different color spaces. Our fusion strategy aims at combining these segmentation maps with a final clustering procedure using as input features, the local histogram of the class labels, previously estimated and associated to each site and for all these initial partitions. This fusion framework remains simple to implement, fast, general enough to be applied to various computer vision applications (e.g., motion detection and segmentation), and has been successfully applied on the Berkeley image database. The experiments herein reported in this paper illustrate the potential of this approach compared to the state-of-the-art segmentation methods recently proposed in the literature.

Keywords: K-means clustering, Berkeley image database, color spaces, fusion of segmentations, textured image segmentation.






M. Mignotte, M. Meunier, J.-P. Soucy.
DCT-based complexity regularization for EM tomographic reconstruction.
IEEE Transactions on Biomedical Engineering, 55(2):801-805,February 2008.

Abstract: This paper introduces a simple algorithm for tomographic reconstruction based on the use of a complexity regularization term. The regularization is formulated in the discrete cosine transform (DCT) domain by promoting a low-noise reconstruction having a high sparsity in the frequency domain. The resulting algorithm simply alternates between a maximum-likelihood (ML) expectation-maximization (EM) update and a decreasing sparsity constraint in the DCT domain. Applications to SPECT reconstruction and comparisons with a classical estimator using the best available regularization terms are given in order to illustrate the potential of our reconstruction technique.

Keywords: Discrete cosine transform (DCT), SPECT tomography, expectation-maximization (EM), reconstruction.





P.-M. Jodoin, M. Mignotte, J. Konrad.
Statistical background subtraction methods using spatial cues.
IEEE Transactions on Circuits and Systems for Video Technology, 17(12):1758-1764, December 2007.

Abstract: Most statistical background subtraction techniques are based on the analysis of temporal color/intensity distribution. However, learning statistics on a series of time frames can be problematic, especially when no frame absent of moving objects is available or when the available memory is not sufficient to store the series of frames needed for learning. In this letter, we propose a spatial variation to the traditional temporal framework. The proposed framework allows statistical motion detection with methods trained on one background frame instead of a series of frames as is usually the case. Our framework includes two spatial background subtraction approaches suitable for different applications. The first approach is meant for scenes having a nonstatic background due to noise, camera jitter or animation in the scene (e.g.,waving trees, fluttering leaves). This approach models each pixel with two PDFs: one unimodal PDF and one multimodal PDF, both trained on one background frame. In this way, the method can handle backgrounds with static and nonstatic areas. The second spatial approach is designed to use as little processing time and memory as possible. Based on the assumption that neighboring pixels often share similar temporal distribution, this second approach models the background with one global mixture of Gaussians.

Keywords: Background detection motion detection.





P.-M. Jodoin, M. Mignotte, C. Rosenberger.
Segmentation framework based on label field fusion.
IEEE Transactions on Image Processing, 16(10):2535-2550, October 2007.

Abstract: In this paper, we put forward a novel fusion framework that mixes together label fields instead of observation data as is usually the case. Our framework takes as input two label fields: a quickly estimated and to-be-refined segmentation map and a spatial region map that exhibits the shape of the main objects of the scene. These two label fields are fused together with a global energy function that is minimized with a deterministic iterative conditional mode algorithm. As explained in the paper, the energy function may implement a pure fusion strategy or a fusion-reaction function. In the latter case, a data-related term is used to make the optimization problem well posed. We believe that the conceptual simplicity, the small number of parameters, the use of a simple and fast deterministic optimizer that admits a natural implementation on a parallel architecture are among the main advantages of our approach. Our fusion framework is adapted to various computer vision applications among which are motion segmentation, motion estimation and occlusion detection.

Keywords: Color segmentation, label fusion, motion estimation, motion segmentation, occlusion.





M. Mignotte.
A post-processing deconvolution step for wavelet-based image denoising methods.
IEEE Signal Processing Letters, 4(9):621-624, September 2007.

Abstract: In this letter, we show that the performance of image denoising algorithms using wavelet transforms can be improved by a post-processing deconvolution step that takes into account the inherent blur function created by the considered wavelet based denoising system. The interest of the proposed deblurring procedure is illustrated on denoised images reconstructed by shrinkage of curvelet and undecimated wavelet coefficients. Experimental results reported here show that the proposed post-processing technique yields improvements in term of image quality and lower mean square error, especially when the image is corrupted by strong additive white Gaussian noise.

Keywords: Curvelet, deblurring, deconvolution, image denoising, nonnegative Garrote shrinkage, undecimated wavelet transform.





F. Destrempes, M. Mignotte, J.-F. Angers.
Localization of shapes using statistical models and stochastic optimization.
IEEE Transactions on Pattern Analysis and Machine Intelligence, 29(9):1603-1615, September 2007.

Abstract: In this paper, we present a new model for deformations of shapes. A pseudolikelihood is based on the statistical distribution of the gradient vector field of the gray level. The prior distribution is based on the probabilistic principal component analysis (PPCA). We also propose a new model based on mixtures of PPCA that is useful in the case of greater variability in the shape. A criterion of global or local object specificity based on a preliminary color segmentation of the image is included into the model. The localization of a shape in an image is then viewed as minimizing the corresponding Gibbs field. We use the exploration/selection (E/S) stochastic algorithm in order to find the optimal deformation. This yields a new unsupervised statistical method for localization of shapes. In order to estimate the statistical parameters for the gradient vector field of the gray level, we use an iterative conditional estimation (ICE) procedure. The color segmentation of the image can be computed with an exploration/selection/estimation (ESE) procedure.

Keywords: Exploration/Selection (E/S) algorithm, Probabilistic Principal Component Analysis (PPCA), Shape localization, statistical model, stochastic optimization.





M. Mignotte.
Image denoising by averaging of piecewise constant simulations of image partitions.
IEEE Transactions on Image Processing, 16(2):523-533, February 2007.

Abstract: This paper investigates the problem of image denoising when the image is corrupted by additive white Gaussian noise. We herein propose a spatial adaptive denoising method which is based on an averaging process performed on a set of Markov Chain Monte-Carlo simulations of region partition maps constrained to be spatially piecewise uniform (i.e., constant in the grey level value sense) for each estimated constant-value regions. For the estimation of these region partition maps, we have adopted the unsupervised Markovian framework in which parameters are automatically estimated in the least square sense. This sequential averaging allows to obtain, under our image model, an approximation of the image to be recovered in the minimal mean square sense error. The experiments reported in this paper demonstrate that the discussed method performs competitively and sometimes better than the best existing state-of-the-art wavelet-based denoising methods in benchmark tests.

Keywords: Image denoising, Markov chain Monte-Carlo (MCMC) simulations, Markovian segmentation.





F. Destrempes, J.-F. Angers, M. Mignotte.
Fusion of hidden Markov Random Field models and its Bayesian estimation.
IEEE Transactions on Image Processing, 15(10):2920-2935, October 2006.

Abstract: In this paper, we present a Hidden Markov Random Field (HMRF) data-fusion model. The proposed model is applied to the segmentation of natural images based on the fusion of colors and textons into Julesz ensembles. The corresponding Exploration/Selection/Estimation (ESE) procedure for the estimation of the parameters is presented. This method achieves the estimation of the parameters of the Gaussian kernels, the mixture proportions, the region labels, the number of regions, and the Markov hyper-parameter. Meanwhile, we present a new proof of the asymptotic convergence of the ESE procedure, based on original finite time bounds for the rate of convergence.

Keywords: Bayesian estimation, Exploration/Selection algorithm, Exploration/Selection/Estimation procedure, Julesz ensembles, Markov Chain Monte Carlo (MCMC) algorithm, color and texture segmentation, fusion of hidden Markov random field models.





P.-M. Jodoin, M. Mignotte.
Markovian segmentation and parameter estimation on graphics hardware.
Journal of Electronic Imaging, 15(3):033005-1-15, 2006.

Abstract: In this paper, we show how Markovian strategies used to solve well-known segmentation problems such as motion estimation, motion detection, motion segmentation, stereovision, and color segmentation can be significantly accelerated when implemented on programmable graphics hardware. More precisely, we expose how the parallel abilities of a standard graphics processing unit usually devoted to image synthesis can be used to infer the labels of a segmentation map. The problems we address are stated in the sense of the maximum a posteriori with an energy-based or probabilistic formulation, depending on the application. In every case, the label field is inferred with an optimization algorithm such as iterated conditional mode (ICM) or simulated annealing. In the case of probabilistic segmentation, mixture parameters are estimated with the K-means and the iterative conditional estimation (ICE) procedure. For both the optimization and the parameter estimation algorithms, the graphics processor unit's (GPU's) fragment processor is used to update in parallel every labels of the segmentation map, while rendering passes and graphics textures are used to simulate optimization iterations. The hardware results obtained with a mid-end graphics card, show that these Markovian applications can be accelerated by a factor of 4 to 200 without requiring any advanced skills in hardware programming.

Keywords:





M. Mignotte.
Segmentation-based regularization term for image deconvolution.
IEEE Transactions on Image Processing, 15(7):1973-1984, July 2006.

Abstract: This paper proposes a new and original inhomogeneous restoration (deconvolution) model under the Bayesian framework for observed images degraded by space-invariant blur and additive Gaussian noise. In this model, regularization is achieved during the iterative restoration process with a segmentation-based a priori term. This adaptive edge-preserving regularization term applies a local smoothness constraint to pre-estimated constant-valued regions of the target image. These constant-valued regions (the segmentation map) of the target image are obtained from a preliminary Wiener deconvolution estimate. In order to estimate reliable segmentation maps, we have also adopted a Bayesian Markovian framework in which the regularized segmentations are estimated in the maximum a posteriori (MAP) sense with the joint use of local Potts prior and appropriate Gaussian conditional luminance distributions. In order to make these segmentations unsupervised, these likelihood distributions are estimated in the maximum likelihood sense. To compute the MAP estimate associated to the restoration, we use a simple steepest descent procedure resulting in an efficient iterative process converging to a globally optimal restoration. The experiments reported in this paper demonstrate that the discussed method performs competitively and sometimes better than the best existing state-of-the-art methods in benchmark tests.

Keywords: Adaptive prior model, Bayesian estimation, Markovian model, Tikhonov regularization,image deconvolution or restoration, image segmentation.






S. Benameur, M. Mignotte, H. Labelle, J.A. De Guise.
A hierarchical statistical modeling approach for the unsupervised 3D biplanar reconstruction of the scoliotic spine.
IEEE Transactions on Biomedical Engineering, 52(12), pp.2041- 2057, December 2005.

Abstract: This paper presents a new and accurate 3D reconstruction technique for the scoliotic spine from a pair of planar and conventional (postero-anterior with normal incidence and lateral) calibrated radiographic images. The proposed model uses a priori hierarchical global knowledge, both on the geometric structure of the whole spine and of each vertebra. More precisely, it relies on the specification of two 3D statistical templates. The first, a rough geometric template on which rigid admissible deformations are defined, is used to ensure a crude registration of the whole spine. An accurate 3D reconstruction is then performed for each vertebra by a second template on which non-linear admissible global, as well as local deformations, are defined. Global deformations are modeled using a statistical modal analysis of the pathological deformations observed on a representative scoliotic vertebra population. Local deformations are represented by a first order Markov process. This unsupervised coarse-to-fine 3D reconstruction procedure leads to two separate minimization procedures efficiently solved in our application with evolutionary stochastic optimization algorithms. In this context, we compare the results obtained with a classical Genetic Algorithm (GA) and a recent Exploration Selection (ES) technique. This latter optimization method with the proposed 3D reconstruction model, is tested on several pairs of biplanar radiographic images with scoliotic deformities. The experiments reported in this paper demonstrate that the discussed method is comparable in terms of accuracy with the classical CT-scan technique while being unsupervised and while requiring only two radiographic images and a lower amount of radiation for the patient.

Keywords: 3D reconstruction model, 3D/2D registration, hierarchical statistical modeling, shape model, biplanar radiographies, scoliosis, medical imaging, energy function minimization, stochastic optimization.






S. Benameur, M. Mignotte, F. Destrempes, J.A. De Guise.
3D biplanar reconstruction of scoliotic rib cage using the estimation of a mixture of probabilistic prior models.
IEEE Transactions on Biomedical Engineering, 52(10):1713-1728, October 2005.

Abstract: In this paper, we present an original method for the 3D reconstruction of the scoliotic rib cage from a planar and a conventional pair of calibrated radiographic images (postero-anterior with normal incidence and lateral). To this end, we first present a robust method for estimating the model parameters in a mixture of Probabilistic Principal Component Analysers (PPCA). This method is based on the Stochastic Expectation Maximization (SEM) algorithm. Parameters of this mixture model are used to constrain the 3D biplanar reconstruction problem of scoliotic rib cage. More precisely, the proposed PPCA mixture model is exploited for dimensionality reduction and to obtain a set of probabilistic prior models associated with each detected class of pathological deformations observed on a representative training scoliotic rib cage population. By using an appropriate likelihood, for each considered class-conditional prior model, the proposed 3D reconstruction is stated as an energy function minimization problem, which is solved with an exploration/selection algorithm. The optimal 3D reconstruction then corresponds to the class of deformation and parameters leading to the minimal energy. This 3D method of reconstruction has been successfully tested and validated on a database of 20 pairs of biplanar radiographic images of scoliotic patients, yielding very promising results. As an alternative to CT-scan 3D reconstruction this scheme has the advantage of low radiation for the patient, and may also be used for diagnosis and evaluation of deformity of a scoliotic rib cage. The proposed method remains sufficiently general to be applied to other reconstruction problems for which a database of objects to be reconstructed is available (with two or more radiographic views).

Keywords: 3D reconstruction model, 3D/2D registration, shape model, reduction of dimensionality, mixtures of probabilistic principal component analysers, biplanar radiographies, scoliosis, medical imaging, stochastic optimization.





F. Destrempes, M. Mignotte, J.-F. Angers.
A stochastic method for Bayesian estimation of Hidden Markov Random Field models with application to a color model.
IEEE Transactions on Image Processing, 14(8):1096-1124, August 2005.

Abstract: We propose a new stochastic algorithm for computing useful Bayesian estimators of hidden Markov random field (HMRF) models that we call exploration/selection/estimation (ESE) procedure. The algorithm is based on an optimization algorithm of O. Francois, called the exploration/selection (E/S) algorithm. The novelty consists of using the a posteriori distribution of the HMRF, as exploration distribution in the E/S algorithm. The ESE procedure computes the estimation of the likelihood parameters and the optimal number of region classes, according to global constraints, as well as the segmentation of the image. In our formulation, the total number of region classes is fixed, but classes are allowed or disallowed dynamically. This framework replaces the mechanism of the split-and-merge of regions that can be used in the context of image segmentation. The procedure is applied to the estimation of a HMRF color model for images, whose likelihood is based on multivariate distributions, with each component following a Beta distribution. Meanwhile, a method for computing the maximum likelihood estimators of Beta distributions is presented. Experimental results performed on 100 natural images are reported. We also include a proof of convergence of the E/S algorithm in the case of nonsymmetric exploration graphs.

Keywords: Bayesian estimation of hidden Markov random field (HMRF) models color model exploration/selection (E/S) algorithm image segmentation maximum likelihood (ML) estimation of Beta distributions.





J.-F. Laliberté, J. Meunier, M. Mignotte, J.P. Soucy.
Detection of diffuse abnormal perfusion in SPECT using a normal brain atlas.
NeuroImage, 23(2):561-8, October 2004.

Abstract: Visual assessment, with significant inter- or intraobserver variability, is still the norm for the evaluation of Single Photon Emission Computerized Tomography (SPECT) cerebral perfusion studies. We present in this paper an automated method for screening SPECT studies to detect diffuse disseminated abnormalities based on a computerized atlas of normal regional cerebral blood flow (rCBF). To generate the atlas, a set of normal brain SPECT studies are registered together. The atlas contains the intensity mean, the nonlinear displacement mean, and the variance of the activity pattern. A patient is then evaluated by registering his or her SPECT volume to the atlas and computing the nonlinear 3-D displacement of each voxel needed for the best shape fit to it. A voxel is counted as "abnormal" if the intensity difference between the atlas and the registered patient (or if the 3-D motion necessary to move the voxel to its registered position) is superior to 3 SD of normal mean. The number of abnormal voxels is used to classify studies. We validated this approach on 24 SPECT perfusion studies selected visually for having clear diffuse anomalies and 21 normal studies. A Markovian segmentation algorithm is also used to identify the white and gray matters for regional analysis. Based on the number of abnormal voxels, two supervised classifiers were tested: (1) minimum distance-to-mean and (2) Bayesian. The analysis of the intensity and displacement "abnormal" voxels allow one to achieve an 80% correct classification rate for the whole brain and a 93% rate if we consider only voxels in the segmented gray matter region.

Keywords: SPECT images, brain atlas, diffuse abnormal perfusion.





F. Destrempes, M. Mignotte.
A statistical model for contours in images.
IEEE Transactions on Pattern Analysis and Machine Intelligence, 26(2):184-197, May 2004.

Abstract: In this paper, we describe a statistical model for the gradient vector field of the gray level in images validated by different experiments. Moreover, we present a global constrained Markov model for contours in images that uses this statistical model for the likelihood. Our model is amenable to an Iterative Conditional Estimation (ICE) procedure for the estimation of the parameters; our model also allows segmentation by means of the Simulated Annealing (SA) algorithm, the Iterated Conditional Modes (ICM) algorithm, or the Modes of Posterior Marginals (MPM) Monte Carlo (MC) algorithm. This yields an original unsupervised statistical method for edge-detection, with three variants. The estimation and the segmentation procedures have been tested on a total of 160 images. Those tests indicate that the model and its estimation are valid for applications that require an energy term based on the log-likelihood ratio. Besides edge-detection, our model can be used for semiautomatic extraction of contours, localization of shapes, non-photo-realistic rendering; more generally, it might be useful in various problems that require a statistical likelihood for contours.

Keywords: Contours in images, edge-detection, parameter estimation, unsupervised statistical segmentation, Markov Random Field model.





M. Mignotte.
Nonparametric multiscale energy-based model and its application in some imagery problems.
IEEE Transactions on Pattern Analysis and Machine Intelligence, 26(2):184-197, February 2004.

Abstract: This paper investigates the use of a nonparametric regularization energy term for devising a example-based rendering and segmentation technique. We have stated this problem in the multiresolution energy minimization framework and exploited the multiscale structure proposed by Wei and Levoy for the texture synthesis problem. In this nonparametric energy minimization framework, we also propose a computationally efficient coarse-to-fine recursive optimization method to minimize the cost function related to this hierarchical model. In this context, the formulation of our example-based regularization term also allows to directly infer an intuitive dissimilarity measure between two contour shapes. This measure is herein exploited to define an efficient shape descriptor for the contour-based shape recognition and indexing problem.

Keywords: Nonparametric multiscale energy-based (or multiresolution example-based) model, inpainting, Non-Photorealistic Rendering (NPR), segmentation, contour-based shape recognition, shape indexing.





S. Benameur, M. Mignotte, S. Parent, H. Labelle, W. Skalli, J. De Guise.
3D/2D registration and segmentation of scoliotic vertebrae using statistical models.
Computerized Medical Imaging and Graphics, July 2003.

Abstract: We propose a new 3D/2D registration method for vertebrae of the scoliotic spine, using two conventional radiographic views (postero-anterior and lateral), and a priori global knowledge of the geometric structure of each vertebra. This geometric knowledge is efficiently captured by a statistical deformable template integrating a set of admissible deformations, expressed by the first modes of variation in Karhunen-Loeve expansion, of the pathological deformations observed on a representative scoliotic vertebra population. The proposed registration method consists of fitting the projections of this deformable template with the preliminary segmented contours of the corresponding vertebra on the two radiographic views. The 3D/2D registration problem is stated as the minimization of a cost function for each vertebra and solved with a gradient descent technique. Registration of the spine is then done vertebra by vertebra. The proposed method efficiently provides accurate 3D reconstruction of each scoliotic vertebra and, consequently, it also provides accurate knowledge of the 3D structure of the whole scoliotic spine. This registration method has been successfully tested on several biplanar radiographic images and validated on 57 scoliotic vertebrae. The validation results reported in this paper demonstrate that the proposed statistical scheme performs better than other conventional 3D reconstruction methods.

Keywords: 3D/2D Registration, 3D reconstruction model, statistical deformable model, shape model, biplanar radiographies, scoliosis, medical imaging, energy function optimization.





M. Mignotte, J. Meunier, J.-P. Soucy, C. Janicki.
Comparison of deconvolution techniques using a distribution mixture parameter estimation: application in spect imagery.
Journal of Electronic Imaging, 11(1), 2002.

Abstract: Thanks to its ability to yield functionally rather than anatomically-based information, the SPECT imagery technique has become a great help in the diagnostic of cerebrovascular diseases which are the third most common cause of death in the USA and Europe. Nevertheless, SPECT images are very blurred and consequently their interpretation is difficult. In order to improve the spatial resolution of these images and then to facilitate their interpretation by the clinician, we propose to implement and to compare the effectiveness of different existing ``blind'' or ``supervised'' deconvolution methods. To this end, we present an accurate distribution mixture parameter estimation procedure which takes into account the diversity of the laws in the distribution mixture of a SPECT image. In our application, parameters of this distribution mixture are efficiently exploited in order to prevent overfitting of the noisy data for the iterative deconvolution techniques without regularization term, or to determine the exact support of the object to be restored when this one is needed. Recent blind deconvolution techniques such as the NAS-RIF algorithm, combined with this estimation procedure, can be efficiently applied in SPECT imagery and yield promising results.

Keywords: SPECT imagery, blind deconvolution, distribution mixture estimation, Markov Random Field model, image restoration





M. Mignotte, J. Meunier, J.-C. Tardif.
Endocardial boundary estimation and tracking in echocardiographic images using deformable templates and Markov Random Fields.
Pattern Analysis and Applications, 4(4):256-271, November 2001.

Abstract: We present a new approach to shape-based segmentation and tracking of deformable anatomical structures in medical images and validate this approach by detecting and tracking the endocardial contour in an echocardiographic image sequence. To this end, some global prior shape knowledges of the endocardial boundary is captured by a prototype template with a set of predefined global and local deformations to take into account its inherent natural variability over time. In this deformable model-based Bayesian segmentation, the data likelihood model relies on an accurate statistical modeling of the grey level distribution of each class present in the ultrasound image. The parameters of this distribution mixture are given by a preliminary iterative estimation step. This estimation scheme relies on a Markov Random Field prior model and takes into account the imaging process as well as the distribution shape of each class present in the image. Then the detection and the tracking problem is stated in a Bayesian framework where it ends up as a cost function minimization problem for each image of the sequence. In our application, this energy optimization problem is efficiently solved by a genetic algorithm combined with a steepest ascent procedure. This technique has been successfully applied on synthetic images and on a real echocardiographic image sequence.

Keywords: Deformable templates, Markov Random Fields, Tracking, Echocardiography, Boundary estimation.





M. Mignotte, J. Meunier.
A multiscale optimization approach for the dynamic contour-based boundary detection issue.
Computerized Medical Imaging and Graphics, 25(3):265-275, May 2001.

Abstract: We present a new multiscale approach for deformable contour optimization. The method relies on a multigrid minimization method and a coarse-to-fine relaxation algorithm. This approach consists in minimizing a cascade of optimization problems of reduced and increasing complexity instead of considering the minimization problem on the full and original configuration space. Contrary to classical multiresolution algorithms, no reduction of image is applied. The family of defined energy functions are derived from the original (full resolution) objective function, ensuring that the same function is handled at each scale and that the energy decreases at each step of the deformable contour minimization process. The efficiency and the speed of this multiscale optimization strategy is demonstrated in the difficult context of the minimization of a region-based contour energy function ensuring the boundary detection of anatomical structures in ultrasound medical imagery. In this context, the proposed multiscale segmentation method is compared to other classical region-based segmentation approaches such as Maximum Likelihood or Markov Random Field-based segmentation techniques. We also extend this multiscale segmentation strategy to active contour models using a classical edge-based likelihood approach. Finally, time and performance analysis of this approach, compared to the (commonly used) dynamic programming-based optimization procedure, is given and allows to attest the accuracy and the speed of the proposed method.

Keywords: active contour model, snake, multiscale optimization, boundary-based segmentation, ultrasound medical images.





K.C. Yao, M. Mignotte, C. Collet, P. Galerne, G. Burel.
Unsupervised segmentation using a SOM and a noise model estimation in sonar imagery.
Pattern Recognition, 33(9):1575-1584, September 2000.

Abstract: This work deals with unsupervised sonar image segmentation. We present a new estimation and segmentation procedure on images provided by a high-resolution sonar. The sonar image is segmented into two kinds of regions: shadow (corresponding to a lack of acoustic reverberation behind each object lying on the seabed) and reverberation (due to the reflection of acoustic wave on the seabed and on the objects). The unsupervised contextual method we propose is defined as a two-step process. Firstly, the iterative conditional estimation is used for the estimation step in order to estimate the noise model parameters and to accurately obtain the proportion of each class in the maximum likelihood sense. Then, the learning of a Kohonen self-organizing map (SOM) is performed directly on the input image to approximate the discriminating functions, i.e. the contextual distribution function of the grey levels. Secondly, the previously estimated proportion, the contextual information and the Kohonen SOM, after learning, are then used in the segmentation step in order to classify each pixel on the input image. This technique has been successfully applied to real sonar images, and is compatible with an automatic processing of massive amounts of data.

Keywords: Kohonen self-organizing map, segmentation, parameter estimation, sonar imagery, Markov random field.





M. Mignotte, C. Collet, P. Pérez, P. Bouthemy.
Markov Random Field and Fuzzy logic modeling in sonar imagery: application to the classification of underwater floor.
Computer Vision and Image Understanding, 79(1):4-24, July 2000.

Abstract: This paper proposes an original method for the classification of seafloors from high resolution sidescan sonar images. We aim at classifying the sonar images into five kinds of regions: sand, pebbles, rocks, ripples, and dunes. The proposed method adopts a pattern recognition approach based on the extraction and the analysis of the cast shadows exhibited by each seabottom type. This method consists of three stages of processing. First, the original image is segmented into two kinds of regions: shadow (corresponding to a lack of acoustic reverberation behind each "object" lying on the seabed) and seabottom reverberation. Second, based on the extracted shadows, shape parameter vectors are computed on subimages and classified with a fuzzy classifier. This preliminary classification is finally refined thanks to a Markov random field model which allows to incorporate spatial homogeneity properties one would expect for the final classification map. Experiments on a variety of real high-resolution sonar images are reported.

Keywords: Sonar imagery, sea-bed classification, shape recognition, Markov Random Field (MRF), fuzzy logic.





M. Mignotte, C. Collet, P. Pérez, P. Bouthemy.
Sonar image segmentation using a hierarchical MRF model.
IEEE Transactions on Image Processing, 9(7):1216-1231, July 2000.

Abstract: This paper is concerned with hierarchical Markov Random Field (MRF) models and their application to sonar image segmentation. We present an original hierarchical segmentation procedure devoted to images given by a high resolution sonar. The sonar image is segmented into two kinds of regions: shadow (corresponding to a lack of acoustic reverberation behind each object lying on the sea-bed) and sea-bottom reverberation. The proposed unsupervised scheme takes into account the variety of the laws in the distribution mixture of a sonar image, and it estimates both the parameters of noise distributions and the parameters of the Markovian prior. For the estimation step, we use an iterative technique which combines a maximum likelihood approach (for noise model parameters) with a least-squares method (for MRF-based prior). In order to model more precisely the local and global characteristics of image content at different scales, we introduce a hierarchical model involving a pyramidal label field. It combines coarse-to-fine causal interactions with a spatial neighborhood structure. This new method of segmentation, called Scale Causal Multigrid (SCM) algorithm, has been successfully applied to real sonar images and seems to be well suited to the segmentation of very noisy images. The experiments reported in this paper demonstrate that the discussed method performs better than other hierarchical schemes for sonar image segmentation.

Keywords: Hierarchical MRF, parameter estimation, sonar imagery, unsupervised segmentation.





M. Mignotte, J. Meunier.
Three-dimensional blind deconvolution of SPECT images.
IEEE Transactions on Biomedical Engineering, 47(2):274-281, February 2000.

Abstract: Thanks to its ability to yield functionally rather than anatomically-based information, the three-dimensional (3D) {SPECT} imagery technique has become a great help in the diagnostic of cerebrovascular diseases. Nevertheless, due to the imaging process, the 3D SPECT images are very blurred and consequently their interpretation by the clinician is often difficult and subjective. In order to improve the resolution of these 3D images and then to facilitate their interpretation, we propose herein, to extend a recent image blind deconvolution technique (called the NAS-RIF deconvolution method) in order to improve both the spatial and the inter-slice resolution of SPECT volumes. This technique requires a preliminary step in order to find the support of the object to be restored. In this paper, we propose to solve this problem with an unsupervised 3D Markovian segmentation technique. This method has been successfully tested on numerous real and simulated brain SPECT volumes, yielding very promising restoration results.

Keywords: SPECT imagery, 3D blind deconvolution, unsupervised segmentation, Markov Random Field model, image restoration.





M. Mignotte, C. Collet, P. Pérez, P. Bouthemy.
Hybrid genetic optimization and statistical model-based approach for the classification of shadow shapes in sonar imagery.
IEEE Transactions on Pattern Analysis and Machine Intelligence, 22(2):129-141, February 2000.

Abstract: We present an original statistical classification method using a deformable template model to separate natural objects from man-made objects in an image provided by a high resolution sonar. A prior knowledge of the manufactured object shadow shape is captured by a prototype template, along with a set of admissible linear transformations, to take into account the shape variability. Then, the classification problem is defined as a two-step process. First, the detection problem of a region of interest in the input image is stated as the minimization of a cost function. Second, the value of this function at convergence allows one to determine whether the desired object is present or not in the sonar image. The energy minimization problem is tackled using relaxation techniques. In this context, we compare the results obtained with a deterministic relaxation technique (a gradient-based algorithm) and two stochastic relaxation methods: Simulated Annealing (SA) and a hybrid Genetic Algorithm (GA). This latter method has been successfully tested on real and synthetic sonar images, yielding very promising results.

Keywords: Deformable template, objective function, simulated annealing, gradient-based algorithm, genetic optimization, shape recognition, sonar imagery.





M. Mignotte, C. Collet, P. Pérez, P. Bouthemy.
Three-class Markovian segmentation of high resolution sonar images.
Computer Vision and Image Understanding, 76(3):191-204, December 1999.

Abstract: This paper presents an original method for analyzing, in an unsupervised way, images supplied by high resolution sonar. We aim at segmenting the sonar image into three kinds of regions: echo areas (due to the reflection of the acoustic wave on the object), shadow areas (corresponding to a lack of acoustic reverberation behind an object lying on the sea-bed), and sea-bottom reverberation areas. This unsupervised method estimates the parameters of noise distributions, modeled by a Weibull probability density function (PDF), and the label field parameters, modeled by a Markov random field (MRF). For the estimation step, we adopt a maximum likelihood technique for the noise model parameters and a least-squares method to estimate the MRF prior model. Then, in order to obtain an accurate segmentation map, we have designed a two-step process that finds the shadow and the echo regions separately, using the previously estimated parameters. First, we introduce a scale-causal and spatial model called SCM (scale causal multigrid), based on a multigrid energy minimization strategy, to find the shadow class. Second, we propose a MRF monoscale model using a priori information (at different level of knowledge) based on physical properties of each region, which allows us to distinguish echo areas from sea-bottom reverberation. This technique has been successfully applied to real sonar images and is compatible with automatic processing of massive amounts of data.

Keywords: sonar imagery, unsupervised segmentation, MRF hierarchical model, Weibull law, noise model estimation.