Guillaume Damiand

A simple paradigm for graph recognition: application to cographs and distance hereditary graphs

An easy way for graph recognition algorithms is to use a two-step process: first, compute a characteristic feature as if the graph belongs to that class; second, check whether the computed characteristic feature as if the graph belongs to that class; second, check whether the computed separating them may yield new and much more easily understood algorithms. In this paper we apply that paradigm to the cograph and distance hereditary graph recognition problems.

Topological model for two-dimensional image representation: definition and optimal extraction algorithm

We define the two-dimensional topological map, a model which represents both topological and geometrical information of a two-dimensional labeled image. Since this model is minimal, complete, and unique, we can use it to define efficient image processing algorithms. The topological map is the last level of a map hierarchy. Each level represents the region boundaries of the image and is defined from the previous one in the hierarchy, thus giving a simple constructive definition. This model is similar to two existing structures but the main innovation of our approach is the progressive definition based on the successive map levels. These different maps can easily be extended in order to define the topological map in any dimension. Furthermore we provide an optimal extraction algorithm which extracts the different maps of the hierarchy in a single image scan. This algorithm is based on local configurations called precodes. Due to our constructive definition, different configurations are factorized which simplifies the implementation.

Removal and Contraction for n-Dimensional Generalized Maps

Removal and contraction are basic operations for several methods conceived in order to handle irregular image pyramids, for multi-level image analysis for instance. Such methods are often based upon graph-like representations which do not maintain all topological information, even for 2-dimensional images. We study the definitions of removal and contraction operations in the generalized maps framework. These combinatorial structures enable us to unambiguously represent the topology of a well-known class of subdivisions of n-dimensional (discrete) spaces. The results of this study make a basis for a further work about irregular pyramids of n-dimensional images.

Topological model for 3D image representation: Definition and incremental extraction algorithm

In this paper, we define the three-dimensional topological map, a model which represents both the topological and geometrical information of a three-dimensional labeled image. Since this model describes the images topology in a minimal way, we can use it to define efficient image processing algorithms. The topological map is the last level of map hierarchy. Each level represents the region boundaries of the image and is defined from the previous level in the hierarchy, thus giving a simple constructive definition. This model is an extension of the similar model defined for 2D images. Progressive definition based on successive map levels allows us to extend this model to higher dimension. Moreover, with progressive definition, we can study each level separately. This simplifies the study of disconnection cases and the proofs of topological map properties. Finally, we provide an incremental extraction algorithm which extracts any map of the hierarchy in a single image scan. Moreover, we show that this algorithm is very efficient by giving the results of our experiments made on artificial images.

Topological Encoding of 3D Segmented Images

In this paper we define the 3d topological map and give an optimal algorithm which computes it from a segmented image. This data structure encodes totally all the information given by the segmentation. More, it allows to continue segmentation either algorithmically or interactively. We propose an original approach which uses several levels of maps. This allows us to propose a reasonable and implementable solution where other approaches dont allow suitable solutions. Moreover our solution has been implemented and the theoretical results translate very well in practical applications.

Split-and-merge algorithms defined on topological maps for 3D image segmentation

Split-and-merge algorithms define a class of image segmentation methods. Topological maps are a mathematical model that represents image subdivisions in 2D and 3D. This paper discusses a split-and-merge method for 3D image data based on the topological map model This model allows representations of states of segmentations and of merge and split operations. Indeed, it can be used as data structure for dynamic changes of segmentation. The paper details such an algorithmic approach and analyzes its time complexity. A general introduction into combinatorial and topological maps is given to support the understanding of the proposed algorithms.

Directly computing the generators of image homology using graph pyramids

We introduce a method for computing homology groups and their generators of a 2D image, using a hierarchical structure, i.e. irregular graph pyramid. Starting from an image, a hierarchy of the image is built by two operations that preserve homology of each region. Instead of computing homology generators in the base where the number of entities (cells) is large, we first reduce the number of cells by a graph pyramid. Then homology generators are computed efficiently on the top level of the pyramid, since the number of cells is small. A top down process is then used to deduce homology generators in any level of the pyramid, including the base level, i.e. the initial image. The produced generators fit on the object boundaries. A unique set of generators called the minimal set, is defined and its computation is discussed. We show that the new method produces valid homology generators and present some experimental results.

Polynomial algorithms for subisomorphism of nD open combinatorial maps

Combinatorial maps describe the subdivision of objects in cells, and incidence and adjacency relations between cells, and they are widely used to model 2D and 3D images. However, there is no algorithm for comparing combinatorial maps, which is an important issue for image processing and analysis. In this paper, we address two basic comparison problems, i.e., map isomorphism, which involves deciding if two maps are equivalent, and submap isomorphism, which involves deciding if a copy of a pattern map may be found in a target map. We formally define these two problems for nD open combinatorial maps, we give polynomial time algorithms for solving them, and we illustrate their interest and feasibility for searching patterns in 2D and 3D images, as any child would aim to do when he searches Wally in Martin Handfords books.

Computing homology group generators of images using irregular graph pyramids

We introduce a method for computing homology groups and their generators of a 2D image, using a hierarchical structure i.e. irregular graph pyramid. Starting from an image, a hierarchy of the image is built, by two operations that preserve homology of each region. Instead of computing homology generators in the base where the number of entities (cells) is large, we first reduce the number of cells by a graph pyramid. Then homology generators are computed efficiently on the top level of the pyramid, since the number of cells is small, and a top down process is then used to deduce homology generators in any level of the pyramid, including the base level i.e. the initial image. We show that the new method produces valid homology generators and present some experimental results.

nD generalized map pyramids: Definition, representations and basic operations

Graph pyramids are often used for representing irregular image pyramids. For the 2D case, combinatorial pyramids have been recently defined in order to explicitly represent more topological information than graph pyramids. The main contribution of this work is the definition of pyramids of n-dimensional (nD) generalized maps. This extends the previous works to any dimension, and generalizes them in order to represent any type of pyramid constructed by using any removal and/or contraction operations. We give basic algorithms that allow to build an nD generalized pyramid that describes a multi-level segmented image. A pyramid of nD generalized maps can be implemented in several ways. We propose three possible representations and give conversion algorithms.