EE-805 / 2 crédits
Remark: July 3-7, 2023 Registration via https://forms.gle/BcQqmwboeWTAmact8
This summer school is an hands-on introduction on the fundamentals of image analysis for scientists. A series of lectures provide students with the key concepts in the field, and are followed by practical sessions with popular software on the participants' own image-analysis software.
The summer school is structured around 7 sessions, each covering a specific imaging topic (see below). There will be one to two sessions per day, each consisting of lectures and interactive practical works on the current thematic:
Fundamentals of Scientific Images
Key concepts in digital imaging (pixel level, contrast, histogram, file format, etc.), multidimensional data (visualisation)
Optics of image formation (light sources, PSF/MTF, etc.), noise, SNR, resolution, 3D scenes, stereo imaging
Operations on Digital Images
Digital filters, morphological operators, segmentation
Optical flow and registration, measuring a displacement field (local, global, discrete), representing and quantifying deformations, tracking of particles (detection and linking)
Machine Learning for Image Analysis
Key terminology, data preparation, existing ML software for image analysis
Deep Learning for Image Analysis
Building and training of models, running of pre-trained models, fine-tuning of pre-trained models
Good Practice for Open Imaging
Ethics of image publication, the do's and don't of figure preparation, storage, formats, licenses, open access, confidentiality
This course is a transversal initiative from the EPFL Center for Imaging.
More information, as well as the schedule are available in this link https://eias-epfl.org/program.
Image analysis, digital images, image acquisition, deep learning, open imagine.
Dans les plans d'études
- Nombre de places: 25
- Forme de l'examen: Ecrit (session libre)
- Matière examinée: Fundamentals of Image Analysis
- Cours: 22 Heure(s)
- Exercices: 9 Heure(s)
- TP: 10 Heure(s)