Slide
Centre Interdisciplinaire
de Recherche et d’Innovation
en Cybersécurité et Société
de Recherche et d’Innovation
en Cybersécurité et Société
1.
Allaoui, M. L.; Allili, M. S.
MEDiXNet: A Robust Mixture of Expert Dermatological Imaging Networks for Skin Lesion Segmentation Article d'actes
Dans: IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recogn., IEEE Computer Society, 2024, ISBN: 19457928 (ISSN); 979-835031333-8 (ISBN), (Journal Abbreviation: IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recogn.).
Résumé | Liens | BibTeX | Étiquettes: Attention mechanism, Attention mechanisms, Blurred boundaries, Cancer detection, Deep learning, Dermatology, Expert systems, Image segmentation, Lesion segmentations, Mixture of experts, Mixture of experts model, Mixture-of-experts model, Salient regions, Skin cancers, Skin lesion, Skin lesion segmentation
@inproceedings{allaoui_medixnet_2024,
title = {MEDiXNet: A Robust Mixture of Expert Dermatological Imaging Networks for Skin Lesion Segmentation},
author = {M. L. Allaoui and M. S. Allili},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85203397643&doi=10.1109%2fISBI56570.2024.10635430&partnerID=40&md5=c95dd2122f03c944e945b684a111e741},
doi = {10.1109/ISBI56570.2024.10635430},
isbn = {19457928 (ISSN); 979-835031333-8 (ISBN)},
year = {2024},
date = {2024-01-01},
booktitle = {IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recogn.},
publisher = {IEEE Computer Society},
abstract = {Skin lesion segmentation in dermatological imaging is crucial for early skin cancer detection. However, it is challenging due to variation in lesion appearance, blurred boundaries, and the presence of artifacts. Existing segmentation methods often fall short in accurately addressing these issues. We present MEDiXNet, a novel deep learning model combining expert networks with the Adaptive Salient Region Attention Module (ASRAM) to specifically tackle these challenges. Tailored for varying lesion types, MEDiXNet leverages ASRAM to enhance focus on critical regions, substantially improving segmentation accuracy. Tested on the ISIC datasets, it achieved a 94% Dice coefficient, surpassing state-of-the-art methods. MEDiXNet's innovative approach represents a significant advancement in dermatological imaging, promising to elevate the precision of skin cancer diagnostics. © 2024 IEEE.},
note = {Journal Abbreviation: IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recogn.},
keywords = {Attention mechanism, Attention mechanisms, Blurred boundaries, Cancer detection, Deep learning, Dermatology, Expert systems, Image segmentation, Lesion segmentations, Mixture of experts, Mixture of experts model, Mixture-of-experts model, Salient regions, Skin cancers, Skin lesion, Skin lesion segmentation},
pubstate = {published},
tppubtype = {inproceedings}
}
Skin lesion segmentation in dermatological imaging is crucial for early skin cancer detection. However, it is challenging due to variation in lesion appearance, blurred boundaries, and the presence of artifacts. Existing segmentation methods often fall short in accurately addressing these issues. We present MEDiXNet, a novel deep learning model combining expert networks with the Adaptive Salient Region Attention Module (ASRAM) to specifically tackle these challenges. Tailored for varying lesion types, MEDiXNet leverages ASRAM to enhance focus on critical regions, substantially improving segmentation accuracy. Tested on the ISIC datasets, it achieved a 94% Dice coefficient, surpassing state-of-the-art methods. MEDiXNet's innovative approach represents a significant advancement in dermatological imaging, promising to elevate the precision of skin cancer diagnostics. © 2024 IEEE.
2.
Nouboukpo, A.; Allaoui, M. L.; Allili, M. S.
Multi-scale spatial consistency for deep semi-supervised skin lesion segmentation Article de journal
Dans: Engineering Applications of Artificial Intelligence, vol. 135, 2024, ISSN: 09521976 (ISSN), (Publisher: Elsevier Ltd).
Résumé | Liens | BibTeX | Étiquettes: Deep learning, Dermatology, Image segmentation, Lesion segmentations, Medical imaging, Multi-scales, Semi-supervised, Semi-supervised learning, Skin lesion, Skin lesion segmentation, Spatial consistency, Spatially constrained mixture model, Spatially-constrained mixture models, Supervised learning, UNets, Unlabeled data
@article{nouboukpo_multi-scale_2024,
title = {Multi-scale spatial consistency for deep semi-supervised skin lesion segmentation},
author = {A. Nouboukpo and M. L. Allaoui and M. S. Allili},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85195700182&doi=10.1016%2fj.engappai.2024.108681&partnerID=40&md5=e1cc2b6a1bb0aed530e8c04583c76167},
doi = {10.1016/j.engappai.2024.108681},
issn = {09521976 (ISSN)},
year = {2024},
date = {2024-01-01},
journal = {Engineering Applications of Artificial Intelligence},
volume = {135},
abstract = {This paper introduces a novel semi-supervised framework, the Multiscale Spatial Consistency Network (MSCNet), for robust semi-supervised skin lesion segmentation. MSCNet uses local and global spatial consistency to leverage a minimal set of labeled data, supplemented by a large number of unlabeled data, to improve segmentation. The model is is based on a single Encoder–Decoder (ED) network, augmented with a Spatially-Constrained Mixture Model (SCMM) to enforce spatial coherence in predictions. To encode the local spatial consistency, a hierarchical superpixel structure is used capture local region context (LRC), bolstering the model capacity to discern fine-grained lesion details. Global consistency is enforced through the SCMM module, which uses a larger context for lesion/background discrimination. In addition, it enables efficient leveraging of the unlabeled data through pseudo-label generation. Experiments demonstrate that the MSCNet outperforms existing state-of-the-art methods in segmenting complex lesions. The MSCNet has an excellent generalization capability, offering a promising direction for semi-supervised medical image segmentation, particularly in scenarios with limited annotated data. The code is available at https://github.com/AdamaTG/MSCNet. © 2024 Elsevier Ltd},
note = {Publisher: Elsevier Ltd},
keywords = {Deep learning, Dermatology, Image segmentation, Lesion segmentations, Medical imaging, Multi-scales, Semi-supervised, Semi-supervised learning, Skin lesion, Skin lesion segmentation, Spatial consistency, Spatially constrained mixture model, Spatially-constrained mixture models, Supervised learning, UNets, Unlabeled data},
pubstate = {published},
tppubtype = {article}
}
This paper introduces a novel semi-supervised framework, the Multiscale Spatial Consistency Network (MSCNet), for robust semi-supervised skin lesion segmentation. MSCNet uses local and global spatial consistency to leverage a minimal set of labeled data, supplemented by a large number of unlabeled data, to improve segmentation. The model is is based on a single Encoder–Decoder (ED) network, augmented with a Spatially-Constrained Mixture Model (SCMM) to enforce spatial coherence in predictions. To encode the local spatial consistency, a hierarchical superpixel structure is used capture local region context (LRC), bolstering the model capacity to discern fine-grained lesion details. Global consistency is enforced through the SCMM module, which uses a larger context for lesion/background discrimination. In addition, it enables efficient leveraging of the unlabeled data through pseudo-label generation. Experiments demonstrate that the MSCNet outperforms existing state-of-the-art methods in segmenting complex lesions. The MSCNet has an excellent generalization capability, offering a promising direction for semi-supervised medical image segmentation, particularly in scenarios with limited annotated data. The code is available at https://github.com/AdamaTG/MSCNet. © 2024 Elsevier Ltd
3.
Messaoudi, H.; Belaid, A.; Allaoui, M. L.; Zetout, A.; Allili, M. S.; Tliba, S.; Salem, D. Ben; Conze, P. -H.
Efficient Embedding Network for 3D Brain Tumor Segmentation Article de journal
Dans: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12658 LNCS, p. 252–262, 2021, ISSN: 03029743, (ISBN: 9783030720834 Publisher: Springer Science and Business Media Deutschland GmbH).
Résumé | Liens | BibTeX | Étiquettes: 3D medical image processing, Brain, Brain tumor segmentation, Classification networks, Convolutional neural networks, Deep learning, Embedding network, Image segmentation, Large dataset, Large datasets, Medical imaging, Natural images, Net networks, Semantic segmentation, Semantics, Signal encoding, Tumors
@article{messaoudi_efficient_2021,
title = {Efficient Embedding Network for 3D Brain Tumor Segmentation},
author = {H. Messaoudi and A. Belaid and M. L. Allaoui and A. Zetout and M. S. Allili and S. Tliba and D. Ben Salem and P. -H. Conze},
editor = {Bakas S. Crimi A.},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85107387134&doi=10.1007%2f978-3-030-72084-1_23&partnerID=40&md5=b3aa3516b0465a1bf5611db4727d95f1},
doi = {10.1007/978-3-030-72084-1_23},
issn = {03029743},
year = {2021},
date = {2021-01-01},
journal = {Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)},
volume = {12658 LNCS},
pages = {252–262},
abstract = {3D medical image processing with deep learning greatly suffers from a lack of data. Thus, studies carried out in this field are limited compared to works related to 2D natural image analysis, where very large datasets exist. As a result, powerful and efficient 2D convolutional neural networks have been developed and trained. In this paper, we investigate a way to transfer the performance of a two-dimensional classification network for the purpose of three-dimensional semantic segmentation of brain tumors. We propose an asymmetric U-Net network by incorporating the EfficientNet model as part of the encoding branch. As the input data is in 3D, the first layers of the encoder are devoted to the reduction of the third dimension in order to fit the input of the EfficientNet network. Experimental results on validation and test data from the BraTS 2020 challenge demonstrate that the proposed method achieve promising performance. © 2021, Springer Nature Switzerland AG.},
note = {ISBN: 9783030720834
Publisher: Springer Science and Business Media Deutschland GmbH},
keywords = {3D medical image processing, Brain, Brain tumor segmentation, Classification networks, Convolutional neural networks, Deep learning, Embedding network, Image segmentation, Large dataset, Large datasets, Medical imaging, Natural images, Net networks, Semantic segmentation, Semantics, Signal encoding, Tumors},
pubstate = {published},
tppubtype = {article}
}
3D medical image processing with deep learning greatly suffers from a lack of data. Thus, studies carried out in this field are limited compared to works related to 2D natural image analysis, where very large datasets exist. As a result, powerful and efficient 2D convolutional neural networks have been developed and trained. In this paper, we investigate a way to transfer the performance of a two-dimensional classification network for the purpose of three-dimensional semantic segmentation of brain tumors. We propose an asymmetric U-Net network by incorporating the EfficientNet model as part of the encoding branch. As the input data is in 3D, the first layers of the encoder are devoted to the reduction of the third dimension in order to fit the input of the EfficientNet network. Experimental results on validation and test data from the BraTS 2020 challenge demonstrate that the proposed method achieve promising performance. © 2021, Springer Nature Switzerland AG.
4.
Nouboukpo, A.; Allili, M. S.
Spatially-coherent segmentation using hierarchical gaussian mixture reduction based on cauchy-schwarz divergence Article de journal
Dans: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11662 LNCS, p. 388–396, 2019, ISSN: 03029743, (ISBN: 9783030272012 Publisher: Springer Verlag).
Résumé | Liens | BibTeX | Étiquettes: Cauchy-Schwarz divergence, Foreground segmentation, Gaussian distribution, Gaussian Mixture Model, Gaussian mixture reduction, Image analysis, Image segmentation, Mixture reductions, Reduction algorithms, Reduction techniques, State-of-art methods
@article{nouboukpo_spatially-coherent_2019,
title = {Spatially-coherent segmentation using hierarchical gaussian mixture reduction based on cauchy-schwarz divergence},
author = {A. Nouboukpo and M. S. Allili},
editor = {Campilho A. Yu A. Karray F.},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85071452890&doi=10.1007%2f978-3-030-27202-9_35&partnerID=40&md5=2689080f7b2410040a038f080ef93bfa},
doi = {10.1007/978-3-030-27202-9_35},
issn = {03029743},
year = {2019},
date = {2019-01-01},
journal = {Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)},
volume = {11662 LNCS},
pages = {388–396},
abstract = {Gaussian mixture models (GMM) are widely used for image segmentation. The bigger the number in the mixture, the higher will be the data likelihood. Unfortunately, too many GMM components leads to model overfitting and poor segmentation. Thus, there has been a growing interest in GMM reduction algorithms that rely on component fusion while preserving the structure of data. In this work, we present an algorithm based on a closed-form Cauchy-Schwarz divergence for GMM reduction. Contrarily to previous GMM reduction techniques which a single GMM, our approach can lead to multiple small GMMs describing more accurately the structure of the data. Experiments on image foreground segmentation demonstrate the effectiveness of our proposed model compared to state-of-art methods. © Springer Nature Switzerland AG 2019.},
note = {ISBN: 9783030272012
Publisher: Springer Verlag},
keywords = {Cauchy-Schwarz divergence, Foreground segmentation, Gaussian distribution, Gaussian Mixture Model, Gaussian mixture reduction, Image analysis, Image segmentation, Mixture reductions, Reduction algorithms, Reduction techniques, State-of-art methods},
pubstate = {published},
tppubtype = {article}
}
Gaussian mixture models (GMM) are widely used for image segmentation. The bigger the number in the mixture, the higher will be the data likelihood. Unfortunately, too many GMM components leads to model overfitting and poor segmentation. Thus, there has been a growing interest in GMM reduction algorithms that rely on component fusion while preserving the structure of data. In this work, we present an algorithm based on a closed-form Cauchy-Schwarz divergence for GMM reduction. Contrarily to previous GMM reduction techniques which a single GMM, our approach can lead to multiple small GMMs describing more accurately the structure of the data. Experiments on image foreground segmentation demonstrate the effectiveness of our proposed model compared to state-of-art methods. © Springer Nature Switzerland AG 2019.
5.
Filali, I.; Allili, M. S.; Benblidia, N.
Multi-graph based salient object detection Article de journal
Dans: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9730, p. 318–324, 2016, ISSN: 03029743, (ISBN: 9783319415000 Publisher: Springer Verlag).
Résumé | Liens | BibTeX | Étiquettes: Graphic methods, Image analysis, Image segmentation, Multi-layer graphs, Multi-scale image decomposition, Multiscale segmentation, Natural images, Object detection, Object recognition, Objective functions, Saliency map, Salient object detection, Salient objects
@article{filali_multi-graph_2016,
title = {Multi-graph based salient object detection},
author = {I. Filali and M. S. Allili and N. Benblidia},
editor = {Karray F. Campilho A. Campilho A.},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84978804496&doi=10.1007%2f978-3-319-41501-7_36&partnerID=40&md5=eb519756d2e72245e4131d5dc0b416b5},
doi = {10.1007/978-3-319-41501-7_36},
issn = {03029743},
year = {2016},
date = {2016-01-01},
journal = {Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)},
volume = {9730},
pages = {318–324},
abstract = {We propose a multi-layer graph based approach for salient object detection in natural images. Starting from a set of multi-scale image decomposition using superpixels, we propose an objective function optimized on a multi-layer graph structure to diffuse saliency from image borders to salient objects. After isolating the object kernel, we enhance the accuracy of our saliency maps through an objectness-like based refinement approach. Beside its simplicity, our algorithm yields very accurate salient objects with clear boundaries. Experiments have shown that our approach outperforms several recent methods dealing with salient object detection. © Springer International Publishing Switzerland 2016.},
note = {ISBN: 9783319415000
Publisher: Springer Verlag},
keywords = {Graphic methods, Image analysis, Image segmentation, Multi-layer graphs, Multi-scale image decomposition, Multiscale segmentation, Natural images, Object detection, Object recognition, Objective functions, Saliency map, Salient object detection, Salient objects},
pubstate = {published},
tppubtype = {article}
}
We propose a multi-layer graph based approach for salient object detection in natural images. Starting from a set of multi-scale image decomposition using superpixels, we propose an objective function optimized on a multi-layer graph structure to diffuse saliency from image borders to salient objects. After isolating the object kernel, we enhance the accuracy of our saliency maps through an objectness-like based refinement approach. Beside its simplicity, our algorithm yields very accurate salient objects with clear boundaries. Experiments have shown that our approach outperforms several recent methods dealing with salient object detection. © Springer International Publishing Switzerland 2016.
6.
Allili, M. S.; Ziou, D.
Likelihood-based feature relevance for figure-ground segmentation in images and videos Article de journal
Dans: Neurocomputing, vol. 167, p. 658–670, 2015, ISSN: 09252312, (Publisher: Elsevier).
Résumé | Liens | BibTeX | Étiquettes: accuracy, algorithm, article, calculation, Feature relevance, Figure-ground segmentations, Gaussian mixture model (GMMs), Image analysis, Image Enhancement, image quality, Image segmentation, Level Set, linear system, mathematical analysis, mathematical model, Negative examples, priority journal, Video cameras, videorecording
@article{allili_likelihood-based_2015,
title = {Likelihood-based feature relevance for figure-ground segmentation in images and videos},
author = {M. S. Allili and D. Ziou},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84952631642&doi=10.1016%2fj.neucom.2015.04.015&partnerID=40&md5=833948d0784e0dc42c2245b9343971dd},
doi = {10.1016/j.neucom.2015.04.015},
issn = {09252312},
year = {2015},
date = {2015-01-01},
journal = {Neurocomputing},
volume = {167},
pages = {658–670},
abstract = {We propose an efficient method for image/video figure-ground segmentation using feature relevance (FR) and active contours. Given a set of positive and negative examples of a specific foreground (an object of interest (OOI) in an image or a tracked objet in a video), we first learn the foreground distribution model and its characteristic features that best discriminate it from its contextual background. For this goal, an objective function based on feature likelihood ratio is proposed for supervised FR computation. FR is then incorporated in foreground segmentation of new images and videos using level sets and energy minimization. We show the effectiveness of our approach on several examples of image/video figure-ground segmentation. © 2015 Elsevier B.V.},
note = {Publisher: Elsevier},
keywords = {accuracy, algorithm, article, calculation, Feature relevance, Figure-ground segmentations, Gaussian mixture model (GMMs), Image analysis, Image Enhancement, image quality, Image segmentation, Level Set, linear system, mathematical analysis, mathematical model, Negative examples, priority journal, Video cameras, videorecording},
pubstate = {published},
tppubtype = {article}
}
We propose an efficient method for image/video figure-ground segmentation using feature relevance (FR) and active contours. Given a set of positive and negative examples of a specific foreground (an object of interest (OOI) in an image or a tracked objet in a video), we first learn the foreground distribution model and its characteristic features that best discriminate it from its contextual background. For this goal, an objective function based on feature likelihood ratio is proposed for supervised FR computation. FR is then incorporated in foreground segmentation of new images and videos using level sets and energy minimization. We show the effectiveness of our approach on several examples of image/video figure-ground segmentation. © 2015 Elsevier B.V.
7.
Boulmerka, A.; Allili, M. Saïd; Ait-Aoudia, S.
A generalized multiclass histogram thresholding approach based on mixture modelling Article de journal
Dans: Pattern Recognition, vol. 47, no 3, p. 1330–1348, 2014, ISSN: 00313203.
Résumé | Liens | BibTeX | Étiquettes: Arbitrary number, Conditional distribution, Gaussian distribution, Gaussian noise (electronic), Generalized Gaussian Distributions, Graphic methods, Histogram thresholding, Image segmentation, Minimum error thresholding, Mixture-modelling, Mixtures, State-of-the-art techniques, Statistical methods, Thresholding, Thresholding methods
@article{boulmerka_generalized_2014,
title = {A generalized multiclass histogram thresholding approach based on mixture modelling},
author = {A. Boulmerka and M. Saïd Allili and S. Ait-Aoudia},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84888328869&doi=10.1016%2fj.patcog.2013.09.004&partnerID=40&md5=d8b872bd0abe9e6c4d52439f8ec360bc},
doi = {10.1016/j.patcog.2013.09.004},
issn = {00313203},
year = {2014},
date = {2014-01-01},
journal = {Pattern Recognition},
volume = {47},
number = {3},
pages = {1330–1348},
abstract = {This paper presents a new approach to multi-class thresholding-based segmentation. It considerably improves existing thresholding methods by efficiently modeling non-Gaussian and multi-modal class-conditional distributions using mixtures of generalized Gaussian distributions (MoGG). The proposed approach seamlessly: (1) extends the standard Otsu's method to arbitrary numbers of thresholds and (2) extends the Kittler and Illingworth minimum error thresholding to non-Gaussian and multi-modal class-conditional data. MoGGs enable efficient representation of heavy-tailed data and multi-modal histograms with flat or sharply shaped peaks. Experiments on synthetic data and real-world image segmentation show the performance of the proposed approach with comparison to recent state-of-the-art techniques. © 2013 Elsevier Ltd. All rights reserved.},
keywords = {Arbitrary number, Conditional distribution, Gaussian distribution, Gaussian noise (electronic), Generalized Gaussian Distributions, Graphic methods, Histogram thresholding, Image segmentation, Minimum error thresholding, Mixture-modelling, Mixtures, State-of-the-art techniques, Statistical methods, Thresholding, Thresholding methods},
pubstate = {published},
tppubtype = {article}
}
This paper presents a new approach to multi-class thresholding-based segmentation. It considerably improves existing thresholding methods by efficiently modeling non-Gaussian and multi-modal class-conditional distributions using mixtures of generalized Gaussian distributions (MoGG). The proposed approach seamlessly: (1) extends the standard Otsu's method to arbitrary numbers of thresholds and (2) extends the Kittler and Illingworth minimum error thresholding to non-Gaussian and multi-modal class-conditional data. MoGGs enable efficient representation of heavy-tailed data and multi-modal histograms with flat or sharply shaped peaks. Experiments on synthetic data and real-world image segmentation show the performance of the proposed approach with comparison to recent state-of-the-art techniques. © 2013 Elsevier Ltd. All rights reserved.
8.
Allili, M. S.
Wavelet modeling using finite mixtures of generalized Gaussian distributions: Application to texture discrimination and retrieval Article de journal
Dans: IEEE Transactions on Image Processing, vol. 21, no 4, p. 1452–1464, 2012, ISSN: 10577149.
Résumé | Liens | BibTeX | Étiquettes: algorithm, Algorithms, article, Automated, automated pattern recognition, computer assisted diagnosis, Computer Simulation, Computer-Assisted, Data Interpretation, Finite mixtures, Generalized Gaussian, Generalized Gaussian Distributions, Image Enhancement, Image Interpretation, Image segmentation, Imaging, Kullback Leibler divergence, Marginal distribution, methodology, Mixtures, Models, Monte Carlo methods, Monte Carlo sampling, Normal Distribution, Pattern Recognition, Performance improvements, reproducibility, Reproducibility of Results, Sensitivity and Specificity, Similarity measure, State-of-the-art approach, Statistical, statistical analysis, statistical model, Texture data set, Texture discrimination, Texture modeling, Textures, three dimensional imaging, Three-Dimensional, Wavelet Analysis, Wavelet coefficients, Wavelet decomposition, Wavelet modeling
@article{allili_wavelet_2012,
title = {Wavelet modeling using finite mixtures of generalized Gaussian distributions: Application to texture discrimination and retrieval},
author = {M. S. Allili},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84859096106&doi=10.1109%2fTIP.2011.2170701&partnerID=40&md5=0420facdc04978ad84bea3126bc1183a},
doi = {10.1109/TIP.2011.2170701},
issn = {10577149},
year = {2012},
date = {2012-01-01},
journal = {IEEE Transactions on Image Processing},
volume = {21},
number = {4},
pages = {1452–1464},
abstract = {This paper addresses statistical-based texture modeling using wavelets. We propose a new approach to represent the marginal distribution of the wavelet coefficients using finite mixtures of generalized Gaussian (MoGG) distributions. The MoGG captures a wide range of histogram shapes, which provides better description and discrimination of texture than using single probability density functions (pdf's), as proposed by recent state-of-the-art approaches. Moreover, we propose a model similarity measure based on Kullback-Leibler divergence (KLD) approximation using Monte Carlo sampling methods. Through experiments on two popular texture data sets, we show that our approach yields significant performance improvements for texture discrimination and retrieval, as compared with recent methods of statistical-based wavelet modeling. © 2011 IEEE.},
keywords = {algorithm, Algorithms, article, Automated, automated pattern recognition, computer assisted diagnosis, Computer Simulation, Computer-Assisted, Data Interpretation, Finite mixtures, Generalized Gaussian, Generalized Gaussian Distributions, Image Enhancement, Image Interpretation, Image segmentation, Imaging, Kullback Leibler divergence, Marginal distribution, methodology, Mixtures, Models, Monte Carlo methods, Monte Carlo sampling, Normal Distribution, Pattern Recognition, Performance improvements, reproducibility, Reproducibility of Results, Sensitivity and Specificity, Similarity measure, State-of-the-art approach, Statistical, statistical analysis, statistical model, Texture data set, Texture discrimination, Texture modeling, Textures, three dimensional imaging, Three-Dimensional, Wavelet Analysis, Wavelet coefficients, Wavelet decomposition, Wavelet modeling},
pubstate = {published},
tppubtype = {article}
}
This paper addresses statistical-based texture modeling using wavelets. We propose a new approach to represent the marginal distribution of the wavelet coefficients using finite mixtures of generalized Gaussian (MoGG) distributions. The MoGG captures a wide range of histogram shapes, which provides better description and discrimination of texture than using single probability density functions (pdf's), as proposed by recent state-of-the-art approaches. Moreover, we propose a model similarity measure based on Kullback-Leibler divergence (KLD) approximation using Monte Carlo sampling methods. Through experiments on two popular texture data sets, we show that our approach yields significant performance improvements for texture discrimination and retrieval, as compared with recent methods of statistical-based wavelet modeling. © 2011 IEEE.
9.
Boulmerka, A.; Allili, M. S.
Thresholding-based segmentation revisited using mixtures of generalized Gaussian distributions Article d'actes
Dans: Proceedings - International Conference on Pattern Recognition, p. 2894–2897, Tsukuba, 2012, ISBN: 978-4-9906441-0-9, (ISSN: 10514651).
Résumé | Liens | BibTeX | Étiquettes: Arbitrary number, Gaussian noise (electronic), Generalized Gaussian Distributions, Heavy-tailed, Image segmentation, Kittler, Minimum error thresholding, Multi-modal, New approaches, Non-Gaussian, Otsu's method, Pattern Recognition, State-of-the-art techniques, Synthetic data
@inproceedings{boulmerka_thresholding-based_2012,
title = {Thresholding-based segmentation revisited using mixtures of generalized Gaussian distributions},
author = {A. Boulmerka and M. S. Allili},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84874575463&partnerID=40&md5=0665cce9aa19af524d1213c1ff728d94},
isbn = {978-4-9906441-0-9},
year = {2012},
date = {2012-01-01},
booktitle = {Proceedings - International Conference on Pattern Recognition},
pages = {2894–2897},
address = {Tsukuba},
abstract = {This paper presents a new approach to image-thresholding-based segmentation. It considerably improves existing methods by efficiently modeling non-Gaussian and multi-modal class-conditional distributions. The proposed approach seamlessly: 1) extends the Otsu's method to arbitrary numbers of thresholds and 2) extends the Kittler and Illingworth minimum error thresholding to non-Gaussian and multi-modal class-conditional data. We use the recently-proposed mixture of generalized Gaussian distributions (MoGG) modeling, which enables to efficiently represent heavy-tailed data, as well as multi-modal histograms with flat and sharply-shaped peaks. Experiments performed on synthetic data and real-world image segmentation show the performance of the proposed approach with comparison to recent state-of-the-art techniques. © 2012 ICPR Org Committee.},
note = {ISSN: 10514651},
keywords = {Arbitrary number, Gaussian noise (electronic), Generalized Gaussian Distributions, Heavy-tailed, Image segmentation, Kittler, Minimum error thresholding, Multi-modal, New approaches, Non-Gaussian, Otsu's method, Pattern Recognition, State-of-the-art techniques, Synthetic data},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper presents a new approach to image-thresholding-based segmentation. It considerably improves existing methods by efficiently modeling non-Gaussian and multi-modal class-conditional distributions. The proposed approach seamlessly: 1) extends the Otsu's method to arbitrary numbers of thresholds and 2) extends the Kittler and Illingworth minimum error thresholding to non-Gaussian and multi-modal class-conditional data. We use the recently-proposed mixture of generalized Gaussian distributions (MoGG) modeling, which enables to efficiently represent heavy-tailed data, as well as multi-modal histograms with flat and sharply-shaped peaks. Experiments performed on synthetic data and real-world image segmentation show the performance of the proposed approach with comparison to recent state-of-the-art techniques. © 2012 ICPR Org Committee.
10.
Larivière, G.; Allili, M. S.
A learning probabilistic approach for object segmentation Article d'actes
Dans: Proceedings of the 2012 9th Conference on Computer and Robot Vision, CRV 2012, p. 86–93, Toronto, ON, 2012, ISBN: 978-076954683-4 (ISBN), (Journal Abbreviation: Proc. Conf. Comput. Rob. Vis., CRV).
Résumé | Liens | BibTeX | Étiquettes: Algorithms, Computer vision, fragments, Image segmentation, Mean shift algorithm, mean-shift algorithm, Object recognition, Object segmentation, Object shape, Optimal segmentation, Probabilistic approaches, Probabilistic Learning, Segmentation process
@inproceedings{lariviere_learning_2012,
title = {A learning probabilistic approach for object segmentation},
author = {G. Larivière and M. S. Allili},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84878376248&doi=10.1109%2fCRV.2012.19&partnerID=40&md5=044a531d9d6de8036a434993f7b5d7ba},
doi = {10.1109/CRV.2012.19},
isbn = {978-076954683-4 (ISBN)},
year = {2012},
date = {2012-01-01},
booktitle = {Proceedings of the 2012 9th Conference on Computer and Robot Vision, CRV 2012},
pages = {86–93},
address = {Toronto, ON},
abstract = {This paper proposes a new method for figure-ground image segmentation based on a probabilistic learning approach of the object shape. Historically, segmentation is mostly defined as a data-driven bottom-up process, where pixels are grouped into regions/objects according to objective criteria, such as region homogeneity, etc. In particular, it aims at creating a partition of the image into contiguous, homogenous regions. In the proposed work, we propose to incorporate prior knowledge about the object shape and category to segment the object from the background. The segmentation process is composed of two parts. In the first part, object shape models are built using sets of object fragments. The second part starts by first segmenting an image into homogenous regions using the mean-shift algorithm. Then, several object hypotheses are tested and validated using the different object shape models as supporting information. As an output, our algorithm identifies the object category, position, as well as its optimal segmentation. Experimental results show the capacity of the approach to segment several object categories. © 2012 IEEE.},
note = {Journal Abbreviation: Proc. Conf. Comput. Rob. Vis., CRV},
keywords = {Algorithms, Computer vision, fragments, Image segmentation, Mean shift algorithm, mean-shift algorithm, Object recognition, Object segmentation, Object shape, Optimal segmentation, Probabilistic approaches, Probabilistic Learning, Segmentation process},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper proposes a new method for figure-ground image segmentation based on a probabilistic learning approach of the object shape. Historically, segmentation is mostly defined as a data-driven bottom-up process, where pixels are grouped into regions/objects according to objective criteria, such as region homogeneity, etc. In particular, it aims at creating a partition of the image into contiguous, homogenous regions. In the proposed work, we propose to incorporate prior knowledge about the object shape and category to segment the object from the background. The segmentation process is composed of two parts. In the first part, object shape models are built using sets of object fragments. The second part starts by first segmenting an image into homogenous regions using the mean-shift algorithm. Then, several object hypotheses are tested and validated using the different object shape models as supporting information. As an output, our algorithm identifies the object category, position, as well as its optimal segmentation. Experimental results show the capacity of the approach to segment several object categories. © 2012 IEEE.