Efficient Matching and Indexing of Graph Models in Content-Based Retrieval

  title={Efficient Matching and Indexing of Graph Models in Content-Based Retrieval},
  author={Stefano Berretti and A. Bimbo and Enrico Vicario},
  journal={IEEE Trans. Pattern Anal. Mach. Intell.},
In retrieval from image databases, evaluation of similarity, based both on the appearance of spatial entities and on their mutual relationships, depends on content representation based on attributed relational graphs. This kind of modeling entails complex matching and indexing, which presently prevents its usage within comprehensive applications. In this paper, we provide a graph-theoretical formulation for the problem of retrieval based on the joint similarity of individual entities and of… 

Genetic approximate matching of attributed relational graphs

A state-of-the-art exhaustive tree search algorithm (Berretti et al., 2001) and a prototype based on a genetic approach are compared, which are universally applicable and do not impose neither any constraints nor any preprocessing steps on the graphs.


The work presented here shows a comparative study between the main methods to build a prototype and the construction of metric trees where the graph prototypes are routing nodes to help to decide the way to explore when the authors make a search in the tree.

Graph indexing: a frequent structure-based approach

The gIndex approach not only provides and elegant solution to the graph indexing problem, but also demonstrates how database indexing and query processing can benefit form data mining, especially frequent pattern mining.

Graph indexing based on discriminative frequent structure analysis

This article proposes a novel indexing model based on discriminative frequent structures that are identified through a graph mining process and shows that the compact index built under this model can achieve better performance in processing graph queries.

Similarity Search in Large-Scale Graph Databases

  • Peixiang Zhao
  • Computer Science
    Handbook of Big Data Technologies
  • 2017
This paper surveys recent graph similarity search techniques and specifically focuses on the work based on the graph edit distance (GED) metric, examining existing GED lower-bounds, graph index structures, and similarity search algorithms in detail, and comparing different methods from multiple aspects including index construction cost, similarity search performance, and applicability in real-world graph databases.

Graph Edit Distance for Active Graph Matching in Content Based Retrieval Applications

A novel model for active graph matching is defined and its application to content based retrieval of images is described and the proposed solution fits with the class of edit distance based techniques and supports active node merging during the graph matching process.

Graph Indexing: Tree + Delta >= Graph

This study verifies that (Tree+Δ) is a better choice than graph for indexing purpose, denoted (Tree-Δ ≥Graph), to address the graph containment query problem and achieves an order of magnitude better performance in index construction.

Vectorization of Structure to Index Graph Databases

This paper addresses graph databases as a domain in-dependent concept and proposes a concept called “structure vectorization” for re-trieval for indexing and retrieval from structure domi-nated graph databases.

Design and evaluation of algorithms for image retrieval by spatial similarity

An algorithm for computing the spatial similarity between two symbolic images that is robust in the sense that it can deal with translation, scale, and rotational variances in images, and has quadratic time complexity in terms of the total number of objects in both the database and query images.

Indexing large metric spaces for similarity search queries

The multivantage point tree structure (mvp-tree) that uses more than one vantage point to partiton the space into spherical cuts at each level is introduced and generalize the idea of using multiple vantage points and the results show that, after all, it may be best to use a large number of vantage points in an internal node in order to end up with a single directory node and keep as many of the precomputed distances as possible to provide more efficient filtering during search operations.

Processing Complex Similarity Queries with Distance-Based Access Methods

This paper considers the relevant case where complex similarity queries are defined through a generic language £ and whose predicates refer to a single feature F and suggests that the index should process complex queries as a whole, thus evaluating multiple similarity predicates at a time.

Weighted walkthroughs between extended entities for retrieval by spatial arrangement

An original framework is presented which supports quantitative nonsymbolic representation and comparison of the mutual positioning of extended nonrectangular spatial entities and properties of the model are expounded to develop an efficient computation technique and to motivate and assess a metric of similarity for quantitative comparison of spatial relationships.

M-tree: An Efficient Access Method for Similarity Search in Metric Spaces

The results demonstrate that the Mtree indeed extends the domain of applicability beyond the traditional vector spaces, performs reasonably well in high-dimensional data spaces, and scales well in case of growing files.

Content-Based Image Indexing

We formulate the content-based image indexing problem as a multi-dimensional nearest-neighbor search problem, and develop/implement an optimistic vantage-point tree algorithm that can dynamically

Hierarchical browsing and search of large image databases

It is shown that the similarity pyramid is best constructed using agglomerative (bottom up) clustering methods, and a fast sparse clustering method which dramatically reduces both memory and computation over conventional methods is presented.

Matching Hierarchical Structures Using Association Graphs

A solution to the problem of matching two trees, by constructing the association graph using the graph-theoretic concept of connectivity, and proving that in the new formulation there is a one-to-one correspondence between maximal cliques and maximal subtree isomorphisms.

Indexing using a spectral encoding of topological structure

An indexing mechanism that maps the topological structure of a tree into a low-dimensional vector space based on a novel eigenvalue characterization of a trees allows us to efficiently retrieve a small set of candidates from a database of models.

Spatial arrangement of color in retrieval by visual similarity