Yves Michels
Doctorant
|
Activities
PhD
Tomographic reconstruction of deformable objects in the heterogeneous ab-initio case
Financing: ANR RHODES.
Director: Mohamed TAJINE.
Supervisor: Loic MAZO.
Supervisor: Etienne BAUDRIER.
This PhD is a part of the RHODES project which the goal is the improvement of the ab-initio tomographic reconstruction methods of deformable objects. This work is focussing on parameters estimations related to the orientation and the state of deformation of a cryo-electron microscopy dataset. The proposed methods are based on non linear dimension reduction and parameters estimation from a scatter plot in a low dimensional space.
keys words - dimension reduction, Isomap, Laplacian graph, Tomography, parameters estimation
Degree project
Video steganalysis based on motion vectors
Supervisor: Ahmed BOURIDANE.
Statistical analysis of the motions vectors used in the H.264 standard to detect steganographic videos. Development of new approach based on the morphology of the motion vector maps. This work were presented in the MIV seminar on January 14th, 2016.
Keys words - H.264, Stéganography, morphological mathematics, SVM
Academic project
High resolution STORM microscopy
Development and implementation of a STORM reconstruction algorithm, in C language.
Keys words - STORM microscopy, Particles detection, Maximum Likelihood, Images matching
Publication
<anyweb> https://icube-publis.unistra.fr/?author=Yves%20Michels </anyweb>
2015
- M.Guillaume, Y.Michels, S.Jay Joint estimation of water column parameters and seabed reflectance combining maximum likelihood and unmixing algorithm 7th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS). June2015