Abstract The extension of mathematical morphology to multivariate data has been an active research topic in recent years. In this paper we propose an approach that relies on the consensus combination of several stochastic permutation orderings. The latter are obtained by searching for a smooth shortest path on a graph representing an image. This path is obtained with a randomized version nearest of neighbors heuristics on a graph. The construction of the graph is of crucial importance and can be based on both spatial and spectral information to enable the obtaining of smoother shortest paths. The starting vertex of a path being taken at random, many different permutation orderings can be obtained and we propose to build a consensus ordering from several permutation orderings. We show the interest of the approach with both quantitative and qualitative results.
{"title":"Mathematical morphology based on stochastic permutation orderings","authors":"O. Lézoray","doi":"10.1515/mathm-2021-0101","DOIUrl":"https://doi.org/10.1515/mathm-2021-0101","url":null,"abstract":"Abstract The extension of mathematical morphology to multivariate data has been an active research topic in recent years. In this paper we propose an approach that relies on the consensus combination of several stochastic permutation orderings. The latter are obtained by searching for a smooth shortest path on a graph representing an image. This path is obtained with a randomized version nearest of neighbors heuristics on a graph. The construction of the graph is of crucial importance and can be based on both spatial and spectral information to enable the obtaining of smoother shortest paths. The starting vertex of a path being taken at random, many different permutation orderings can be obtained and we propose to build a consensus ordering from several permutation orderings. We show the interest of the approach with both quantitative and qualitative results.","PeriodicalId":244328,"journal":{"name":"Mathematical Morphology - Theory and Applications","volume":"91 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124213056","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Fraser Macfarlane, P. Murray, S. Marshall, B. Perret, A. Evans, Henry White
Abstract The extension of Mathematical Morphology to colour and multivariate images is challenging due to the need to define a total ordering in the colour space. No one general way of ordering multivariate data exists and, therefore, there is no single, definitive way of performing morphological operations on colour images. In this paper, we propose an extension to mathematical morphology, based on reduced ordering, specifically the morphological Hit-or-Miss Transform which is used for object detection. The reduced ordering employed transforms multivariate observations to scalar comparisons allowing for an order to be derived and for both flat and non-flat structuring elements to be used. We also compare other definitions of the Hit-or-Miss Transform and test alternative colour ordering schemes presented in the literature. Our proposed method is shown to be intuitive and outperforms other approaches to multivariate Hit-or-Miss Transforms. Furthermore, methods of setting the parameters of the proposed Hit-or-Miss Transform are introduced in order to make the transform robust to noise and partial occlusion of objects and, finally, a set of design tools are presented in order to obtain optimal values for setting these parameters accordingly.
{"title":"Robust Object Detection in Colour Images Using a Multivariate Percentage Occupancy Hit-or-Miss Transform","authors":"Fraser Macfarlane, P. Murray, S. Marshall, B. Perret, A. Evans, Henry White","doi":"10.1515/mathm-2020-0111","DOIUrl":"https://doi.org/10.1515/mathm-2020-0111","url":null,"abstract":"Abstract The extension of Mathematical Morphology to colour and multivariate images is challenging due to the need to define a total ordering in the colour space. No one general way of ordering multivariate data exists and, therefore, there is no single, definitive way of performing morphological operations on colour images. In this paper, we propose an extension to mathematical morphology, based on reduced ordering, specifically the morphological Hit-or-Miss Transform which is used for object detection. The reduced ordering employed transforms multivariate observations to scalar comparisons allowing for an order to be derived and for both flat and non-flat structuring elements to be used. We also compare other definitions of the Hit-or-Miss Transform and test alternative colour ordering schemes presented in the literature. Our proposed method is shown to be intuitive and outperforms other approaches to multivariate Hit-or-Miss Transforms. Furthermore, methods of setting the parameters of the proposed Hit-or-Miss Transform are introduced in order to make the transform robust to noise and partial occlusion of objects and, finally, a set of design tools are presented in order to obtain optimal values for setting these parameters accordingly.","PeriodicalId":244328,"journal":{"name":"Mathematical Morphology - Theory and Applications","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116565964","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract Flat morphology is a general method for obtaining increasing operators on grey-level or multivalued images from increasing operators on binary images (or sets). It relies on threshold stacking and superposition; equivalently, Boolean max and min operations are replaced by lattice-theoretical sup and inf operations. In this paper we consider the construction a flat operator on grey-level or colour images from an operator on binary images that is not increasing. Here grey-level and colour images are functions from a space to an interval in ℝm or ℤm (m ≥ 1). Two approaches are proposed. First, we can replace threshold superposition by threshold summation. Next, we can decompose the non-increasing operator on binary images into a linear combination of increasing operators, then apply this linear combination to their flat extensions. Both methods require the operator to have bounded variation, and then both give the same result, which conforms to intuition. Our approach is very general, it can be applied to linear combinations of flat operators, or to linear convolution filters. Our work is based on a mathematical theory of summation of real-valued functions of one variable ranging in a poset. In a second paper, we will study some particular properties of non-increasing flat operators.
{"title":"Flat morphological operators from non-increasing set operators, I: general theory","authors":"C. Ronse","doi":"10.1515/mathm-2020-0109","DOIUrl":"https://doi.org/10.1515/mathm-2020-0109","url":null,"abstract":"Abstract Flat morphology is a general method for obtaining increasing operators on grey-level or multivalued images from increasing operators on binary images (or sets). It relies on threshold stacking and superposition; equivalently, Boolean max and min operations are replaced by lattice-theoretical sup and inf operations. In this paper we consider the construction a flat operator on grey-level or colour images from an operator on binary images that is not increasing. Here grey-level and colour images are functions from a space to an interval in ℝm or ℤm (m ≥ 1). Two approaches are proposed. First, we can replace threshold superposition by threshold summation. Next, we can decompose the non-increasing operator on binary images into a linear combination of increasing operators, then apply this linear combination to their flat extensions. Both methods require the operator to have bounded variation, and then both give the same result, which conforms to intuition. Our approach is very general, it can be applied to linear combinations of flat operators, or to linear convolution filters. Our work is based on a mathematical theory of summation of real-valued functions of one variable ranging in a poset. In a second paper, we will study some particular properties of non-increasing flat operators.","PeriodicalId":244328,"journal":{"name":"Mathematical Morphology - Theory and Applications","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134361754","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract We propose a novel efficient seed-based method for the multi-object segmentation of images based on graphs, named Hierarchical Layered Oriented Image Foresting Transform (HLOIFT). It uses a tree of the relations between the image objects, with each node in the tree representing an object. Each tree node may contain different individual high-level priors of its corresponding object and defines a weighted digraph, named as layer. The layer graphs are then integrated into a hierarchical graph, considering the hierarchical relations of inclusion and exclusion. A single energy optimization is performed in the hierarchical layered weighted digraph leading to globally optimal results satisfying all the high-level priors. The experimental evaluations of HLOIFT, on medical, natural, and synthetic images, indicate promising results comparable to the related baseline methods that include structural information, but with lower computational complexity. Compared to the hierarchical segmentation by the min-cut/max-flow algorithm, our approach is less restrictive, leading to globally optimal results in more general scenarios, and has a better running time.
{"title":"Efficient Hierarchical Multi-Object Segmentation in Layered Graphs","authors":"L. C. Leon, K. Ciesielski, P. A. Miranda","doi":"10.1515/mathm-2020-0108","DOIUrl":"https://doi.org/10.1515/mathm-2020-0108","url":null,"abstract":"Abstract We propose a novel efficient seed-based method for the multi-object segmentation of images based on graphs, named Hierarchical Layered Oriented Image Foresting Transform (HLOIFT). It uses a tree of the relations between the image objects, with each node in the tree representing an object. Each tree node may contain different individual high-level priors of its corresponding object and defines a weighted digraph, named as layer. The layer graphs are then integrated into a hierarchical graph, considering the hierarchical relations of inclusion and exclusion. A single energy optimization is performed in the hierarchical layered weighted digraph leading to globally optimal results satisfying all the high-level priors. The experimental evaluations of HLOIFT, on medical, natural, and synthetic images, indicate promising results comparable to the related baseline methods that include structural information, but with lower computational complexity. Compared to the hierarchical segmentation by the min-cut/max-flow algorithm, our approach is less restrictive, leading to globally optimal results in more general scenarios, and has a better running time.","PeriodicalId":244328,"journal":{"name":"Mathematical Morphology - Theory and Applications","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121581123","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract In many domains of information processing, such as knowledge representation, preference modeling, argumentation, multi-criteria decision analysis, spatial reasoning, both vagueness, or imprecision, and bipolarity, encompassing positive and negative parts of information, are core features of the information to be modeled and processed. This led to the development of the concept of bipolar fuzzy sets, and of associated models and tools, such as fusion and aggregation, similarity and distances, mathematical morphology. Here we propose to extend these tools by defining algebraic and topological relations between bipolar fuzzy sets, including intersection, inclusion, adjacency and RCC relations widely used in mereotopology, based on bipolar connectives (in a logical sense) and on mathematical morphology operators. These definitions are shown to have the desired properties and to be consistent with existing definitions on sets and fuzzy sets, while providing an additional bipolar feature. The proposed relations can be used for instance for preference modeling or spatial reasoning. They apply more generally to any type of functions taking values in a poset or a complete lattice, such as L-fuzzy sets.
{"title":"Modeling Imprecise and Bipolar Algebraic and Topological Relations using Morphological Dilations","authors":"I. Bloch","doi":"10.1515/mathm-2020-0107","DOIUrl":"https://doi.org/10.1515/mathm-2020-0107","url":null,"abstract":"Abstract In many domains of information processing, such as knowledge representation, preference modeling, argumentation, multi-criteria decision analysis, spatial reasoning, both vagueness, or imprecision, and bipolarity, encompassing positive and negative parts of information, are core features of the information to be modeled and processed. This led to the development of the concept of bipolar fuzzy sets, and of associated models and tools, such as fusion and aggregation, similarity and distances, mathematical morphology. Here we propose to extend these tools by defining algebraic and topological relations between bipolar fuzzy sets, including intersection, inclusion, adjacency and RCC relations widely used in mereotopology, based on bipolar connectives (in a logical sense) and on mathematical morphology operators. These definitions are shown to have the desired properties and to be consistent with existing definitions on sets and fuzzy sets, while providing an additional bipolar feature. The proposed relations can be used for instance for preference modeling or spatial reasoning. They apply more generally to any type of functions taking values in a poset or a complete lattice, such as L-fuzzy sets.","PeriodicalId":244328,"journal":{"name":"Mathematical Morphology - Theory and Applications","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115945229","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Behzad Mirmahboub, D. S. Maia, François Merciol, S. Lefèvre
Abstract Representing an image through a tree structure as provided with a morphological hierarchy enables efficient image analysis and processing methods operating directly on the tree structure. Max-tree and min-tree can be built with efficient algorithms but they only focus on brighter and darker components of the image respectively. Conversely, the Tree-of-Shapes is a self-complementary image representation that provides access to all regional extrema of the image (both brighter and darker components), but its computation is more time-consuming. In this paper, we introduce a new, simple and efficient tree structure called median-tree. It relies on a median image that is straightforwardly constructed by subtracting the median pixel value from an image to decompose it into positive and negative parts. The median tree can then be obtained by applying the efficient max-tree algorithms available in the literature on this median image. We show through theoretical and experimental studies that the median-tree offers similar characteristics to the Tree-of-Shapes, but comes with a considerably lower construction complexity.
{"title":"Median-Tree: An Efficient Counterpart of Tree-of-Shapes","authors":"Behzad Mirmahboub, D. S. Maia, François Merciol, S. Lefèvre","doi":"10.1515/mathm-2020-0110","DOIUrl":"https://doi.org/10.1515/mathm-2020-0110","url":null,"abstract":"Abstract Representing an image through a tree structure as provided with a morphological hierarchy enables efficient image analysis and processing methods operating directly on the tree structure. Max-tree and min-tree can be built with efficient algorithms but they only focus on brighter and darker components of the image respectively. Conversely, the Tree-of-Shapes is a self-complementary image representation that provides access to all regional extrema of the image (both brighter and darker components), but its computation is more time-consuming. In this paper, we introduce a new, simple and efficient tree structure called median-tree. It relies on a median image that is straightforwardly constructed by subtracting the median pixel value from an image to decompose it into positive and negative parts. The median tree can then be obtained by applying the efficient max-tree algorithms available in the literature on this median image. We show through theoretical and experimental studies that the median-tree offers similar characteristics to the Tree-of-Shapes, but comes with a considerably lower construction complexity.","PeriodicalId":244328,"journal":{"name":"Mathematical Morphology - Theory and Applications","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125400505","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract When fitting archaeological artifacts, one would like to have a representation that simplifies fragments while preserving their complementarity. In this paper, we propose to employ the scale-spaces of mathematical morphology to hierarchically simplify potentially fitting fracture surfaces. We study the masking effect when morphological operations are applied to selected subsets of objects. Since fitting locally depends on the complementarity of fractures only, we introduce ‘Boundary Morphology’ on surfaces rather than volumes. Moreover, demonstrating the Lipschitz nature of the terracotta fractures informs our novel extrusion method to compute both closing and opening operations simultaneously. We also show that in this proposed representation the effects of abrasion and uncertainty are naturally bounded, justifying the morphological approach. This work is an extension of our contribution earlier published in the proceedings of ISMM2019 [10].
{"title":"Boundary Morphology for Hierarchical Simplification of Archaeological Fragments","authors":"H. ElNaghy, L. Dorst","doi":"10.1515/mathm-2020-0101","DOIUrl":"https://doi.org/10.1515/mathm-2020-0101","url":null,"abstract":"Abstract When fitting archaeological artifacts, one would like to have a representation that simplifies fragments while preserving their complementarity. In this paper, we propose to employ the scale-spaces of mathematical morphology to hierarchically simplify potentially fitting fracture surfaces. We study the masking effect when morphological operations are applied to selected subsets of objects. Since fitting locally depends on the complementarity of fractures only, we introduce ‘Boundary Morphology’ on surfaces rather than volumes. Moreover, demonstrating the Lipschitz nature of the terracotta fractures informs our novel extrusion method to compute both closing and opening operations simultaneously. We also show that in this proposed representation the effects of abrasion and uncertainty are naturally bounded, justifying the morphological approach. This work is an extension of our contribution earlier published in the proceedings of ISMM2019 [10].","PeriodicalId":244328,"journal":{"name":"Mathematical Morphology - Theory and Applications","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128719042","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract In this article, we present a fast-marching based algorithm for generating superpixel (FMS) partitions of images. The idea behind the algorithm is to draw an analogy between waves propagating in a heterogeneous medium and regions growing on an image at a rate depending on the local color and texture. The FMS algorithm is evaluated on the Berkeley Segmentation Dataset 500. It yields results in terms of boundary adherence that are slightly better than the ones obtained with similar approaches including the Simple Linear Iterative Clustering, the Eikonal-based region growing for efficient clustering and the Iterative Spanning Forest framework for superpixel segmentation algorithms. An interesting feature of the proposed algorithm is that it can take into account texture information to compute the superpixel partition. We illustrate the interest of adding texture information on a specific set of images obtained by recombining textures patches extracted from images representing stripes, originally constructed by Giraud et al. [20]. On this dataset, our approach works significantly better than color based superpixel algorithms.
{"title":"Fast marching based superpixels","authors":"Kaiwen Chang, B. Figliuzzi","doi":"10.1515/mathm-2020-0105","DOIUrl":"https://doi.org/10.1515/mathm-2020-0105","url":null,"abstract":"Abstract In this article, we present a fast-marching based algorithm for generating superpixel (FMS) partitions of images. The idea behind the algorithm is to draw an analogy between waves propagating in a heterogeneous medium and regions growing on an image at a rate depending on the local color and texture. The FMS algorithm is evaluated on the Berkeley Segmentation Dataset 500. It yields results in terms of boundary adherence that are slightly better than the ones obtained with similar approaches including the Simple Linear Iterative Clustering, the Eikonal-based region growing for efficient clustering and the Iterative Spanning Forest framework for superpixel segmentation algorithms. An interesting feature of the proposed algorithm is that it can take into account texture information to compute the superpixel partition. We illustrate the interest of adding texture information on a specific set of images obtained by recombining textures patches extracted from images representing stripes, originally constructed by Giraud et al. [20]. On this dataset, our approach works significantly better than color based superpixel algorithms.","PeriodicalId":244328,"journal":{"name":"Mathematical Morphology - Theory and Applications","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125457600","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
R. Biswas, Gaëlle Largeteau-Skapin, R. Zrour, Eric Andres
Abstract Rhombic dodecahedron is a space filling polyhedron which represents the close packing of spheres in 3D space and the Voronoi structures of the face centered cubic (FCC) lattice. In this paper, we describe a new coordinate system where every 3-integer coordinates grid point corresponds to a rhombic dodecahedron centroid. In order to illustrate the interest of the new coordinate system, we propose the characterization of 3D digital plane with its topological features, such as the interrelation between the thickness of the digital plane and the separability constraint we aim to obtain. We also present the characterization of 3D digital lines and study it as the intersection of multiple digital planes. Characterization of 3D digital sphere with relevant topological features is proposed as well along with the 48-symmetry appearing in the new coordinate system.
{"title":"Digital Objects in Rhombic Dodecahedron Grid","authors":"R. Biswas, Gaëlle Largeteau-Skapin, R. Zrour, Eric Andres","doi":"10.1515/mathm-2020-0106","DOIUrl":"https://doi.org/10.1515/mathm-2020-0106","url":null,"abstract":"Abstract Rhombic dodecahedron is a space filling polyhedron which represents the close packing of spheres in 3D space and the Voronoi structures of the face centered cubic (FCC) lattice. In this paper, we describe a new coordinate system where every 3-integer coordinates grid point corresponds to a rhombic dodecahedron centroid. In order to illustrate the interest of the new coordinate system, we propose the characterization of 3D digital plane with its topological features, such as the interrelation between the thickness of the digital plane and the separability constraint we aim to obtain. We also present the characterization of 3D digital lines and study it as the intersection of multiple digital planes. Characterization of 3D digital sphere with relevant topological features is proposed as well along with the 48-symmetry appearing in the new coordinate system.","PeriodicalId":244328,"journal":{"name":"Mathematical Morphology - Theory and Applications","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127052121","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Samy Blusseau, Bastien Ponchon, S. Velasco-Forero, J. Angulo, I. Bloch
Abstract This paper addresses the issue of building a part-based representation of a dataset of images. More precisely, we look for a non-negative, sparse decomposition of the images on a reduced set of atoms, in order to unveil a morphological and explainable structure of the data. Additionally, we want this decomposition to be computed online for any new sample that is not part of the initial dataset. Therefore, our solution relies on a sparse, non-negative auto-encoder, where the encoder is deep (for accuracy) and the decoder shallow (for explainability). This method compares favorably to the state-of-the-art online methods on two benchmark datasets (MNIST and Fashion MNIST) and on a hyperspectral image, according to classical evaluation measures and to a new one we introduce, based on the equivariance of the representation to morphological operators.
{"title":"Approximating morphological operators with part-based representations learned by asymmetric auto-encoders","authors":"Samy Blusseau, Bastien Ponchon, S. Velasco-Forero, J. Angulo, I. Bloch","doi":"10.1515/mathm-2020-0102","DOIUrl":"https://doi.org/10.1515/mathm-2020-0102","url":null,"abstract":"Abstract This paper addresses the issue of building a part-based representation of a dataset of images. More precisely, we look for a non-negative, sparse decomposition of the images on a reduced set of atoms, in order to unveil a morphological and explainable structure of the data. Additionally, we want this decomposition to be computed online for any new sample that is not part of the initial dataset. Therefore, our solution relies on a sparse, non-negative auto-encoder, where the encoder is deep (for accuracy) and the decoder shallow (for explainability). This method compares favorably to the state-of-the-art online methods on two benchmark datasets (MNIST and Fashion MNIST) and on a hyperspectral image, according to classical evaluation measures and to a new one we introduce, based on the equivariance of the representation to morphological operators.","PeriodicalId":244328,"journal":{"name":"Mathematical Morphology - Theory and Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129953501","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：