Frédéric J.P. Richard

Frédéric J.P. Richard

Professor of Applied Mathematics

Aix-Marseille Université

Biography

Frédéric Richard is a professor of applied mathematics at Aix-Marseille university, member of the Institute of Mathematics of Marseille. His current interests are at the crossing of image processing, spatial statistics and data science. He designs and studies mathematical models and methods for the statistical analysis of structured data such as images, surfaces, or graphs. He applies his methods in different domains including medicine, neuroscience and industry.

For twenty years, Frédéric Richard has been teaching mathematics, statistics and programming languages (python, matlab, sas) at all graduate levels of the university. He is particularly involved in the program Data Science of the master’s degree Applied Mathematics, Statistics. He has led its implementation and organization between 2017 and 2020. Frédéric Richard also supervises master internships and PhD students.

Beyond his teaching activity, Frédéric Richard is the director of the department of mathematics, managing the mathematical degrees of the university.

Besides, Frédéric Richard leads actions to promote and ease the interactions between academic laboratories and private companies. From 2012 to 2020, he was a regional officer for the Agency for Mathematics in Interaction with Companies and Society (AMIES). He is now the director of the service Eureka at the Institute Archimede.

Interests
  • Image processing
  • Spatial statistics
  • Data science
Education

  • French research habilitation, 2009

    Université de Paris (Paris Descartes)

  • PhD Thesis in Applied Mathematics, 2000

    Université de Paris (Paris Descartes)

  • MS Mathematics, Vision, Machine Learning, 1997

    Ecole Normale Supérieure Paris Saclay

Experience

 
 
 
 
 
Aix-Marseille University.
Professor of Applied Mathematics
Aix-Marseille University.
Sep 2011 – Present Marseille

Responsibilities include:

  • Director of the master’s degree of applied maths, stats (2012-2020).
  • Head of the team Signal and Image at I2M (2012-2020).
 
 
 
 
 
AMU, UFR Sciences.
Director of Department of mathematics.
AMU, UFR Sciences.
Jan 2021 – Present Marseille
Management of the university’s math training.
 
 
 
 
 
AMU, Institute Archimède.
Director of Eureka (part-time).
AMU, Institute Archimède.
Jan 2020 – Present Marseille
Promotion of interactions between academic laboratories and private companies.
 
 
 
 
 
AMIES.
Regional officier (part-time).
AMIES.
Jan 2012 – Jan 2020 Marseille
Organization of events for promoting interactions between academic laboratories and private companies.
 
 
 
 
 
University Paris-Descartes
Associate Professor
University Paris-Descartes
Sep 2001 – Aug 2011 Paris

Responsibilities include:

  • PI of a multidisciplinary project (ANR and INCA fundings) on the topic of the statistical analysis of medical image textures (MATAIM project).
  • Member of Scientific Council and Board of Directors of Paris Descartes University under the chairmanship of Pr. Axel Kahn

Recent Posts

Preprint: Full inference for the anisotropic fractional Brownian field.
The anisotropic fractional Brownian field (AFBF) is a non-stationary Gaussian random field which has been used for the modeling of textured images. In this paper, we address the open issue of estimating the functional parameters of this field, namely the topothesy and Hurst functions.
Last updated on Jan 10, 2023 1 min read
Project
Publication: PyAFBF: a Python library for sampling image textures from the anisotropic fractional Brownian field.
The Python library PyAFBF is devoted to the simulation of anisotropic textures of image. These textures are sampled from a mathematical model called the anisotropic fractional Brownian field (AFBF) (Bonami & Estrade, 2003); see Figure 1 for an illustration.
Richard
Last updated on Jul 20, 2022 1 min read
Project
Publication: FDG-PET to T1 Weighted MRI Translation with 3D Elicit Generative Adversarial Network (E-GAN)
With the strengths of deep learning, computer-aided diagnosis (CAD) is a hot topic for researchers in medical image analysis. One of the main requirements for training a deep learning model is providing enough data for the network.
Farideh Bazangani; Frédéric Richard; Baddih Ghattas; Eric Guedj
Last updated on Jan 4, 2023 1 min read
Research talk : "Full inference for the anisotropic fractional Brownian field" (joint work with Paul Escande).
The anisotropic fractional Brownian field is a non-stationary Gaussian field (Bonami and Estrade, 2003) which has been used for the modeling of image microtextures (Richard, 2016-18). Having stationary increments, its probability distribution is characterized by a semi-variogram whose spectral representation in polar coordinates
Last updated on May 27, 2022 2 min read
Inauguration d'Eureka
Eureka est une instance de l’Institut Archimède qui s’adresse aux entreprises à la recherche d’une expertise en mathématique et informatique pour le développement de technologies et l’innovation. Il fait partie du réseau MSO ("Modéliser, Simuler, Optimiser") animé par l'AMIES (agence maths-entreprises).
Last updated on Dec 29, 2021 1 min read

Recent & Upcoming Talks

Talk "Full inference for the anisotropic fractional Brownian field" (joint work with Paul Escande).

The anisotropic fractional Brownian field is a non-stationary Gaussian field (Bonami and Estrade, 2003) which has been used for the …
Jun 7, 2022 5:06 PM — Jun 9, 2022 10:00 PM Laboratoire Paul Painlevé, Villeneuve d'Ascq
Project
Anisotropic fractional Brownian fields: properties, simulation and estimation.
In this talk, we focus on the anisotropic fractional Brownian field (AFBF), which is a multi-dimensional extension of the fractional …
Dec 7, 2021 — Dec 8, 2021 Blida (Algeria) - Visioconference.
Project
Inauguration d'Eureka
Eureka est une instance de l’Institut Archimède qui s’adresse aux entreprises à la recherche d’une expertise en …
Dec 6, 2021 — Riality lab Palais de la Bourse 9 La Canebière 13001 Marseille FRANCE

Publications

For a complete list of publications, please consult one of the following links:

HAL Repository / Google Scholar.

Codes and Softwares

Please visit the Github repositories.

Projects

*
All Data science Statistics Image processing
80-Prime
The project is part of a global interdisciplinary context concerning brain molecular imaging. The objective is to develop statistical methodologies for the classification of textures using a Bayesian approach which combines random field modeling with Deep Convolutional Neural Networks.
80-Prime
OrganoDyn
Reconstruction de dynamiques cellulaires à travers les échelles: comment les mouvements tissulaires construisent-ils les formes ? Collaboration avec Philippe Roudot avec le soutien de l’Institut CENTURI. Encadrement de la thèse de Jules Venaret.
OrganoDyn
PyAFBF.
Synthesis and analysis of image textures through anisotropic fractional Brownian fields.
PyAFBF.
BIFROST [2018-2021].
Blind Identification, Filtering and Restoration On Spectral Techniques.

Collaborators

Caroline Chaux

Aix-Marseille University - CNRS

CNRS Researcher

Eric Guedj

APHM

Professeur Université Practicien Hospitalier

Julien Lefèvre

Institut des Neurosciences de La Timone

Associate Professor of Computer Science

Olivier Coulon

Institut des Neurosciences de la Timone

Researcher

Paul Escande

Aix-Marseille University

CNRS Researcher

Philippe Roudot

Aix-Marseille University - Institute Centuri

Researcher Chaire Century

Actual PhD Students

Avatar

Farideh Baganzani

Aix-Marseille Unniversity

PhD Student

deep learning, image processing, medical imaging

Jules Vanaret

Aix-Marseille University

PhD student

Former PhD Students

Benjamin Coulaud

Aix-Marseille université

Consultant

statistics, applied mathematics

Françoise Tilotta

APHP

MD - PHD

forensic science, image processing

Avatar

Hamed Rabiei

Data Scientist at Neurochlore

image processing, medical imaging

Huong Vu

Aix-Marseille Université

Consultant

statistics, applied mathematics

Avatar

Khuyen Le

Postdoctoral Researcher at 3IA Côte d’Azur

statistics, deep learning, medical imaging

Avatar

Mohammed Hachama

Professor at University of Blida 1, Vice-Rector

image processing, applied mathematics

Courses

Enseignements dispensés en 2021-22.

  • Statistiques, Licence Mathématiques, Physique, Chimie, Informatique, 3e année.
  • Optimisation pour les sciences de données, Master mathématiques appliquées, statistique, Parcours Data Science, 1ère année.
  • IA et Image, Master mathématiques appliquées, statistique, Parcours Data Science, 2e année.

Ebooks (en construction).

  • Statistique inférentielle (licence).
  • Optimisation pour les sciences de données

Statistique inférentielle en Python

Présentation

Les notebooks en python qui sont proposés ci-dessous s’adressent à des étudiants de licence scientifique qui souhaitent assimiler par la pratique les notions vues en cours de statistique inférentielle. En créant un compte kaggle, ils peuvent être réalisés directement sur un navigateur web sans avoir à installer python. Vous pouvez également proposer des corrections en ligne et les partager. Bon travail !

Pré-requis

  • Notions de base en probabilité et statistique inférentielle.
  • Notions de base en programmation python.

Note

Le cours en ligne qui accompagne ces notebooks est en construction.

Commencer Consulter le cours

Statistique inférentielle en Python

Contact