Keywords

Integer Linear Programming Automated entomology Appariement de graphes Symbol recognition Image quality assessment Graph matching Graph Matching Graph Edit Distance Local Branching Heuristic Bipartite graph matching Bag of Words Arthropods Approximation poly Graph distance Pattern Recognition Classification Attributed graphs Pattern Matching Cadastre Algorithmes génétiques
› Full list

Researcher identifiers

Number of documents

60

Structural machine learning and discrete optimization for graph matching and graph classification

Romain Raveaux: In 2006, he obtained two M.Sc., one in networking and telecommunication and another one in computer sciences from the University of Rouen (France). From 2006 to 2010, he had worked on his Ph.D. Thesis at the L3I laboratory of the University of La Rochelle. Since 2011, he is an Associate Professor at the computer sciences lab LIFAT of the University of Tours. His work concerns the structural pattern recognition, proposing contributions related respectively to supervised graph classification and graph matching. Especially, a focus is given on similarity and dissimilarity computation on graphs. His work is applied to image analysis problems for object recognition and detection as well as molecule classification (Chemoinformatics). His scientific course is at the confluence of two research areas: combinatorial optimization and machine learning. Results and contributions of its research works have already been published in the forms of 3 theses and more than 40 research papers including 15 International Journals and 25 International conferences and workshops. Along his research activities, he has built strong collaboration with other research laboratories/universities at national and international levels. Accordingly, he has an active collaboration with 3 national laboratories/universities in France and one international collaboration with an Australian laboratory. As a lecturer, teaching assistant and associate professor, he has conducted teaching at different universities and institutions. He was a lecturer for the International Master of Research in Computer Science: Computer Aided and Decision Support at the University of Tours.

Keywords:

  • Computer vision
  • Machine Learning
  • Structural Pattern Recognition
  • Graph matching
  • Graph neural network
  • Discrete optimization
  • Reinforcement learning / Transfert learning

 My research work in three presentations:

  1. Graph matching and classification from combinatorial optimization and machine learning viewpoints (PDF).
  2. Biodiversity estimation trough insect image classification (PDF). 
  3. Document image quality assessment (PDF).

 

Contact : romain.raveaux@univ-tours.fr

I – Research Activity

 I am a researcher at the Computer Science laboratory of Tours (LIFAT).

A – Center of Interest

  • Computer Vision (Detection, Recognition, Tracking)
  • Machine learning (Learning theory, Mathematical modelling, Graph neural network, Reinforcement learning)
  • Discrete optimization (Mathematical programming, Branch and bound, Matheuristics)

B - Research projects (2 topics)

Topic : Environnement

CARAMBA project (Regional project 2016-2019) : Entomology has had many applications in many biological domains (i.e insect counting as a biodiversity index). To meet a growing biological demand and to compensate a decreasing workforce amount, automated entomology has been around for decades. This challenge has been tackled by computer scientists as well as by biologists themselves. Views are adopted on image capture, feature extraction, classification methods and the tested datasets. 

Tasks: 

  • Image capture
  • Image recognition

Publication : 2 journal papers and 2 conference papers.

Presentation : French poster (Orasis 2017) and  Presentation in English : Effective Training of Convolutional Neural Networks for Insect Image Recognition.

 

Topic: Digital humanities and documents

 

VISIT project : In 2019, I was involved in setting up a research project of regional interest called VISIT. This project has been accepted. A post-doc will work for a year and a half on the development of an incremental algorithm for recognizing and indexing geo-localized images. In collaboration with the castle of Loche .

 

DOD project : in 2014-2015, I was in charge with Donatello Conte (LIFAT) of a work package named Document Image Quality Assessment of a project called Document Image Compression in relation with ITSOFT company.

Publication : 2 journal papers and 2 conference papers. 

Presentation : Estimating quality of documents (2015). ITESOFT Aimargues.

 

ALPAGE: From 2006 to 2010, the framework of my thesis was part of the ALPAGE project. Diachronic AnaLyse of Urban Space PArisian: GEomatic approach: ALPAGE is a research program, initiated in 2006, thanks to the support of the National Research Agency. Based on the association of 4 laboratories, and the collaboration of many other partners (Historian, Geomatics, …). 

 

II - Teaching 

I am a teacher at Polytech’Tours engineering school of the university of Tours.

Networking: 127 h per year (Level L3) from 2013 to 2019.

Mobile programming: 36 h per year (Level M2) from 2013 to 2019.

Multimedia systems: 18 h per year (Level M2) from 2013 to 2019.

Python and data science: 38 h per year (Level L2) from 2016 to 2019.

System programming: 12 h per year (Level L3) shell, fork, shared memory, pthread, semaphore. From 2012 to 2014.

Pattern recognition: 8h per year (Level M2) from 2015 to 2017.

Industrial Networking: 12h per year. CAN protocol (Level M1) from 2015 to 2017.

C for microcontroller: 10h per year (Level M1) from 2015 to 2017.

 

 More demos and materials on my website.

Journal articles18 documents

  • Mostafa Darwiche, Donatello Conte, Romain Raveaux, Vincent T'Kindt. Graph Edit Distance: Accuracy of Local Branching from an application point of view. Pattern Recognition Letters, Elsevier, In press, ⟨10.1016/j.patrec.2018.03.033⟩. ⟨hal-01761595⟩
  • Zeina Abu-Aisheh, Romain Raveaux, Jean-Yves Ramel. Efficient k-nearest neighbors search in graph space. Pattern Recognition Letters, Elsevier, 2018, ⟨10.1016/j.patrec.2018.05.001⟩. ⟨hal-01791560⟩
  • Maxime Martineau, Romain Raveaux, Donatello Conte, Gilles Venturini. Learning error-correcting graph matching with a multiclass neural network. Pattern Recognition Letters, Elsevier, 2018, ⟨10.1016/j.patrec.2018.03.031⟩. ⟨hal-01758990⟩
  • Mostafa Darwiche, Donatello Conte, Romain Raveaux, Vincent T'Kindt. A local branching heuristic for solving a Graph Edit Distance Problem. Computers & Operations Research, 2018, ⟨10.1016/j.cor.2018.02.002⟩. ⟨hal-01587928⟩
  • Alireza Alaei, Donatello Conte, Maxime Martineau, Romain Raveaux. Blind Document Image Quality Prediction Based on Modification of Quality Aware Clustering Method Integrating a Patch Selection Strategy. Expert Systems with Applications, Elsevier, 2018, ⟨10.1016/j.eswa.2018.05.007⟩. ⟨hal-01792116⟩
  • Zeina Abu-Aisheh, Romain Raveaux, Jean-Yves Ramel, Patrick Martineau. A parallel graph edit distance algorithm. Expert Systems with Applications, Elsevier, 2018, 94, pp.41 - 57. ⟨10.1016/j.eswa.2017.10.043⟩. ⟨hal-01629290⟩
  • Julien Lerouge, Zeina Abu-Aisheh, Romain Raveaux, Pierre Héroux, Sébastien Adam. New binary linear programming formulation to compute the graph edit distance. Pattern Recognition, Elsevier, 2017, 72, pp.254 - 265. ⟨10.1016/j.patcog.2017.07.029⟩. ⟨hal-01619313⟩
  • Zeina Abu-Aisheh, Benoit Gaüzère, Sébastien Bougleux, Jean-Yves Ramel, Luc Brun, et al.. Graph edit distance contest: Results and future challenges. Pattern Recognition Letters, Elsevier, 2017, 100, pp.96-103. ⟨10.1016/j.patrec.2017.10.007⟩. ⟨hal-01624592⟩
  • Maxime Martineau, Donatello Conte, Romain Raveaux, Ingrid Arnault, Damien Munier, et al.. A survey on image-based insect classification. Pattern Recognition, Elsevier, 2017, 65, pp.273 - 284. ⟨10.1016/j.patcog.2016.12.020⟩. ⟨hal-01441203⟩
  • Julien Lerouge, Zeina Abu-Aisheh, Romain Raveaux, Pierre Héroux, Sébastien Adam. New binary linear programming formulation to compute the graph edit distance. Pattern Recognition, Elsevier, 2017, 72, pp.254 - 265. ⟨10.1016/j.patcog.2017.07.029⟩. ⟨hal-01619308⟩
  • Zeina Abu-Aisheh, Romain Raveaux, Jean-Yves Ramel. Anytime graph matching. Pattern Recognition Letters, Elsevier, 2016, 84, pp.215--224. ⟨10.1016/j.patrec.2016.10.004⟩. ⟨hal-01490832⟩
  • Samuel Leturcq, Romain Raveaux. Les graphes pour étudier les dynamiques spatiales à partir des séries fiscales médiévales et modernes. Etat des lieux de l’expérience Modelespace. Bulletin du Centre d'études médiévales d'Auxerre. Hors-série, Centre d'études médiévales d'Auxerre ; ARTEHIS, UMR CNRS/uB, 2016, Géolocalisation et sources anciennes ? Actes des journées d'études de Dijon, Maison des Sciences de l'Homme, 13-14 novembre 2014, ⟨https://cem.revues.org/13770⟩. ⟨10.4000/cem.13805⟩. ⟨hal-01311067⟩
  • Alireza Alaei, Romain Raveaux, Donatello Conte. Image quality assessment based on regions of interest. Signal, Image and Video Processing, Springer Verlag, 2016, 11 (4), pp.673-680. ⟨10.1007/s11760-016-1009-z⟩. ⟨hal-01405149⟩
  • Romain Raveaux, Jean-Christophe Burie, Jean-Marc Ogier. Structured Representations in a Content Based Image Retrieval Context. Journal of Visual Communication and Image Representation, Elsevier, 2013, 24 (8), pp.1252-1268. ⟨10.1016/j.jvcir.2013.08.010⟩. ⟨hal-00936497⟩
  • Romain Raveaux, J.-C. Burie, Jean-Marc Ogier. Structured representations in a content based image retrieval context. Journal of Visual Communication and Image Representation, Elsevier, 2013, Volume 24 (Issue 8), pp.1252-1268. ⟨10.1016/j.jvcir.2013.08.010⟩. ⟨hal-01079013⟩
  • Romain Raveaux, Jean-Christophe Burie, Jean-Marc Ogier. A local evaluation of vectorized documents by means of polygon assignments and matching. International Journal on Document Analysis and Recognition, Springer Verlag, 2012, 15 (1), pp.21-43. ⟨10.1007/s10032-010-0143-3⟩. ⟨hal-00717341⟩
  • Romain Raveaux, Sébastien Adam, Pierre Héroux, Éric Trupin. Learning graph prototypes for shape recognition. Computer Vision and Image Understanding, Elsevier, 2011, 115 (7), pp.905 - 918. ⟨10.1016/j.cviu.2010.12.015⟩. ⟨hal-00593453⟩
  • Romain Raveaux, Jean-Christophe Burie, Jean-Marc Ogier. A graph matching method and a graph matching distance based on subgraph assignments. Pattern Recognition Letters, Elsevier, 2009, pp.In Press, Corrected Proof. ⟨10.1016/j.patrec.2009.10.011⟩. ⟨hal-00442028⟩

Conference papers35 documents

  • Mostafa Darwiche, Romain Raveaux, Donatello Conte, Vincent T'Kindt. Résoudre le problème de la distance d'édition entre graphes avec les matheuristiques. ROADEF19, Feb 2019, Le Havre, France. ⟨hal-02077682⟩
  • Mostafa Darwiche, Donatello Conte, Romain Raveaux, Vincent T'Kindt. Solving the graph edit distance problem with variable partitioning local search. IAPR International workshop on Graph-Based Representation in Pattern Recognition (GbR19), Jun 2019, Tours, France. ⟨hal-01579501⟩
  • Alireza Alaei, Romain Raveaux, Donatello Conte, Bela Stantic. "Quality" vs. "Readability" in Document Images: Statistical Analysis of Human Perception. 2018 13th IAPR International Workshop on Document Analysis Systems (DAS), Apr 2018, Vienna, Austria. ⟨10.1109/DAS.2018.42⟩. ⟨hal-01879782⟩
  • Maxime Martineau, Romain Raveaux, Clément Chatelain, Donatello Conte, Gilles Venturini. Effective Training of Convolutional Neural Networks for Insect Image Recognition. 19th International Conference on Advanced Concepts for Intelligent Vision Systems, Sep 2018, Poitiers, France. ⟨hal-01887615⟩
  • Mostafa Darwiche, Donatello Conte, Romain Raveaux, Vincent T'Kindt. Formulation linéaire en nombres entiers pour le problème de la distance d'édition entre graphes. ROADEF18, Feb 2018, Lorient, France. ⟨hal-01717264⟩
  • Mostafa Darwiche, Romain Raveaux, Donatello Conte, Vincent T'Kindt. A New Mixed Integer Linear Program for the Graph Edit Distance Problem. ISCO18, Apr 2018, Marrakesh, Morocco. pp.254 - 265. ⟨hal-01717268⟩
  • Rafael Werneck, Romain Raveaux, Salvatore Tabbone, Ricardo da S Torres. Learning Cost Functions for Graph Matching. Joint IAPR International Workshops on Statistical Techniques in Pattern Recognition (SPR) and Structural and Syntactic Pattern Recognition (SSPR), Aug 2018, Beijing, China. ⟨hal-01889964⟩
  • Mostafa Darwiche, Romain Raveaux, Donatello Conte, Vincent T'Kindt. Solving a special case of the Graph Edit Distance Problem with Local Branching. Matheuristics 2018, Jun 2018, Tours, France. ⟨hal-01717709⟩
  • Mostafa Darwiche, Romain Raveaux, Donatello Conte, Vincent T'Kindt. Graph Edit Distance in the exact context. Joint IAPR International Workshops on Statistical Techniques in Pattern Recognition (SPR) and Structural and Syntactic Pattern Recognition (SSPR), Aug 2018, Bejing, China. pp. 326-336. ⟨hal-01880093⟩
  • Mostafa Darwiche, Donatello Conte, Romain Raveaux, Vincent T'Kindt. Evaluation de modèles mathématiques pour le problème de la distance d'édition entre graphes. ROADEF2017, Feb 2017, Metz, France. ⟨hal-01564064⟩
  • Mostafa Darwiche, Donatello Conte, Romain Raveaux, Vincent T'Kindt. The Graph Edit Distance Problem treated by the Local Branching Heuristic. MIC17 12th Metaheuristics International Conference, Jul 2017, Barcelona, Spain. ⟨hal-01564079⟩
  • Zeina Abu-Aisheh, Romain Raveaux, Jean-Yves Ramel. Fast Nearest Neighbors Search in Graph Space based on a Branch-and-Bound Strategy. Graph-Based Representations in Pattern Recognition - 11th IAPR-TC-15 International Workshop, GbRPR, May 2017, Anacapri, Italy. ⟨hal-01617383⟩
  • Romain Raveaux, Maxime Martineau, Donatello Conte, Gilles Venturini. Learning Graph Matching with a Graph-Based Perceptron in a Classification Context. International Workshop on Graph-Based Representations in Pattern Recognition GbRPR 2017, May 2017, Anacapri, Italy. ⟨hal-01576056⟩
  • Alireza Alaei, Donatello Conte, Michael Blumenstein, Romain Raveaux. Document Image Quality Assessment Based on Texture Similarity Index. 12th IAPR Workshop on Document Analysis Systems (DAS), 2016, Santorini, Greece. pp.132-137, ⟨10.1109/DAS.2016.33⟩. ⟨hal-01429042⟩
  • Julien Lerouge, Zeina Abu-Aisheh, Romain Raveaux, Pierre Héroux, Sébastien Adam. Exact Graph Edit Distance Computation Using a Binary Linear Program. Structural, Syntactic, and Statistical Pattern Recognition - Joint IAPR International Workshop, S+SSPR 2016, Mérida, Mexico, November 29 - December 2, 2016, Proceedings, 2016, Unknown, Unknown Region. pp.485--495, ⟨10.1007/978-3-319-49055-7_43⟩. ⟨hal-01491027⟩
  • Zeina Abu-Aisheh, Romain Raveaux, Jean-Yves Ramel. A Graph Database Repository and Performance Evaluation Metrics for Graph Edit Distance. Graph-Based Representations in Pattern Recognition - 10th IAPR-TC-15 International Workshop, GbRPR 2015, Beijing, China, May 13-15, 2015. Proceedings, May 2015, Beijing, France. ⟨hal-01168809⟩
  • Zeina Abu-Aisheh, Romain Raveaux, Jean-Yves Ramel, Patrick Martineau. An Exact Graph Edit Distance Algorithm for Solving Pattern Recognition Problems. 4th International Conference on Pattern Recognition Applications and Methods 2015, Jan 2015, Lisbon, Portugal. ⟨10.5220/0005209202710278⟩. ⟨hal-01168816⟩
  • Alireza Alaei, Donatello Conte, Romain Raveaux. Document Image Quality Assessment based on Improved Gradient Magnitude Similarity Deviation. 13th International Conference on Document Analysis and Recognition, Aug 2015, Nancy, France. ⟨hal-01224875⟩
  • Ernest Valveny, Matthieu Delalandre, Romain Raveaux, Bart Lamiroy. Report on the Symbol Recognition and Spotting Contest. 9th International Workshop, GREC 2011, Seoul, Korea, September 15-16, 2011, Revised Selected Papers, Sep 2011, Seoul, North Korea. pp.198-207. ⟨hal-01178076⟩
  • Mickaël Coustaty, Romain Raveaux, Jean-Marc Ogier. Historical document analysis: A review of French projects and open issues. EUSIPCO 2011, Aug 2011, Barcelona, Spain. pp.1445 - 1449. ⟨hal-01247944⟩
  • Mickaël Coustaty, Romain Raveaux, Jean-Marc Ogier. Historical Document Analysis : A Review of French Projects and Open Issues. EUSIPCO 2011, Aug 2011, Barcelona, Spain. ⟨hal-00916024⟩
  • Romain Raveaux, Jean-Christophe Burie, Jean-Marc Ogier. Object Extraction from Colour Cadastral Maps. IAPR International Workshop on Document Analysis Systems, Sep 2008, Japan. pp.506-514. ⟨hal-00385910⟩
  • Romain Raveaux, Eugen Barbu, Sébastien Adam, Pierre Héroux, Éric Trupin. Graphes prototypes vs. graphe médian généralisé pour la classification de données structurées. Colloque International Francophone sur l'Ecrit et le Document, Oct 2008, France. pp.37-42. ⟨hal-00335037⟩
  • Romain Raveaux, Jean-Christophe Burie, Jean-Marc Ogier. A colour text/graphics separation based on a graph representation. International Conference on Pattern Recognition (ICPR), Dec 2008, Tampa, United States. pp.1-4. ⟨hal-00385906⟩
  • Romain Raveaux, Jean-Christophe Burie, Jean-Marc Ogier. A Colour Space Selection Scheme dedicated to Information Retrieval Tasks. Pattern Recognition in Information Systems, Proceedings of the 8th International Workshop on Pattern Recognition in Information Systems, PRIS 2008, In conjunction with ICEIS 2008, Barcelona, Spain, June 2008, Jun 2008, Barcelone, Spain. pp.123-134. ⟨hal-00385923⟩
  • Romain Raveaux, Eugen Barbu, Hervé Locteau, Sébastien Adam, Pierre Héroux, et al.. A Graph Classification Approach Using a Multi-objective Genetic Algorithm Application to Symbol Recognition. In Graph-Based Representations in Pattern Recognition, 6th IAPR-TC-15 International Workshop GbRPR 2007, Lecture Notes in Computer Science, Francisco Escolano and Mario Vento Eds., Springer, ISBN: 978-3-540-72902-0, 2007, France. pp.361-370. ⟨hal-00361099⟩
  • Romain Raveaux, Jean-Christophe Burie, Jean-Marc Ogier. A knowledge-based colour segmentation scheme. Application to ancient cadastral maps. Seventh IAPR International Workshop on Graphics Recognition - (GREC 2007), 2007, France. pp.69-71. ⟨hal-00361100⟩
  • Romain Raveaux, Jean-Christophe Burie, Jean-Marc Ogier. A contribution to Ancient Cadastral Maps interpretation through color analysis. The 7th International Workshop on Pattern Recognition in Information Systems (PRIS 2007), 2007, France. pp.n.a. ⟨hal-00361101⟩
  • Romain Raveaux, Jean-Christophe Burie, Jean-Marc Ogier. A colour document interpretation: Application to ancient cadastral maps. International conference on document analysis and recognition (ICDAR 2007), 2007, France. pp.116-120. ⟨hal-00361102⟩
  • Eugen Barbu, Romain Raveaux, Hervé Locteau, Sébastien Adam, Pierre Héroux, et al.. Graph Classification Using Genetic Algorithm and Graph Probing Application to Symbol Recognition. ICPR (3), 2006, Hong Kong SAR China. pp.296-299. ⟨hal-00440176⟩
  • Hervé Locteau, Romain Raveaux, Sébastien Adam, Yves Lecourtier, Pierre Héroux, et al.. Approximation of Digital Curves using a Multi-Objective Genetic Algorithm. ICPR (2), 2006, Hong Kong SAR China. pp.716-719. ⟨hal-00440172⟩
  • Hervé Locteau, Romain Raveaux, Sébastien Adam, Yves Lecourtier, Pierre Héroux, et al.. Approximation de courbes par Algorithme Génétique Multi-Objectif. Colloque International Francophone sur l'Ecrit et le Document, Sep 2006, France. pp.37-42. ⟨hal-00113904⟩
  • Eugen Barbu, Romain Raveaux, Hervé Locteau, Sébastien Adam, Pierre Héroux, et al.. Classification de graphes par algorithmes génétiques et signatures de graphes. Application à la reconnaissance de symboles. Colloque International Francophone sur l'Ecrit et le Document, Sep 2006, France. pp.91-96. ⟨hal-00113906⟩
  • Hervé Locteau, Romain Raveaux, Sebastien Adam, Yves Lecourtier, Pierre Héroux, et al.. Approximation de courbes par Algorithme Génétique Multi-Objectif. CIFED 2006, Sep 2006, France. pp.151-156. ⟨hal-00134802⟩
  • Hervé Locteau, Romain Raveaux, Sébastien Adam, Yves Lecourtier, Pierre Héroux, et al.. Polygonal Approximation of Digital Curves Using a Multi-objective Genetic Algorithm. IAPR International Workshop on Graphics Recognition, 2005, Honk-Kong, Hong Kong SAR China. pp.300-311, ⟨10.1007/11767978_27⟩. ⟨hal-00601529⟩

Book sections4 documents

  • Mostafa Darwiche, Romain Raveaux, Donatello Conte, Vincent T'Kindt. A Local Branching Heuristic for the Graph Edit Distance Problem. Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications. CIARP 2017, 10657, pp.194-202, 2018, 978-3-319-75193-1. ⟨10.1007/978-3-319-75193-1_24⟩. ⟨hal-02077687⟩
  • Ernest Valveny, Mathieu Delalandre, Romain Raveaux, Bart Lamiroy. Report on the Symbol Recognition and Spotting Contest. Young-Bin Kwon and Jean-Marc Ogier. Graphics Recognition. New Trends and Challenges. 9th International Workshop, GREC 2011, Seoul, Korea, September 15-16, 2011, Revised Selected Papers, 7423, Springer, pp.198-207, 2013, Lecture Notes in Computer Science, 978-3-642-36823-3. ⟨10.1007/978-3-642-36824-0_19⟩. ⟨hal-00788250⟩
  • Romain Raveaux, Jean-Christophe Burie, Jean-Marc Ogier. La vectorisation automatisée des plans Vasserot, dans Paris de parcelles en pixels. Analyse géomatique de l'espace parisien médiéval et moderne, Presse Universitaire de Vincennes, pp.54-65, 2013, 978-2-84292-364-8. ⟨hal-00936555⟩
  • Romain Raveaux, Jean-Christophe Burie, Jean-Marc Ogier. A Segmentation Scheme Based on a Multi-graph Representation: Application to Colour Cadastral Maps. Graphics Recognition. Recent Advances and New Opportunities, Springer Berlin / Heidelberg, pp.202-212, 2008, Lecture Notes in Computer Science, ⟨10.1007/978-3-540-88188-9_20⟩. ⟨hal-00442029⟩

Preprints, Working Papers, ...3 documents

  • Julien Lerouge, Zeina Abu-Aisheh, Romain Raveaux, Pierre Héroux, Sébastien Adam. Graph edit distance : a new binary linear programming formulation. 2017. ⟨hal-01619358⟩
  • Zeina Abu-Aisheh, Romain Raveaux, Jean-Yves Ramel, Patrick Martineau. A Parallel Graph Edit Distance Algorithm. 2017. ⟨hal-01476393⟩
  • Romain Raveaux, Jean-Christophe Burie, Jean-Marc Ogier. A graph matching method based on probe assignments. 2008. ⟨hal-00305232v3⟩