HAH802E1M Sensors and image processing (4 ECTS)
Presentation
Description:
The aim of this module is to enable students to acquire the basic knowledge required in most of the disciplines associated with engineering sciences concerning digital images and their use. To follow this unit, students must have some knowledge of linear algebra, as well as some notions of continuous and sampled signal processing. As practical work is carried out in Matlab, some programming skills in this language are desirable. On completion of this course, students will have mastered the basics of image processing. They will understand the principles on which most image analysis and modification software is based.
Hourly volumes :
CM: 16.5 h
TD: 12 h
TP: 12 h
Field: 0 h
Prerequisites :
L3 in science or health.
Recommended prerequisites :
R.A.S
More information
Assessment of knowledge :
Written final exam with second session (coefficient 0.6) + Practical work exam on report (coefficient 0.4).
Syllabus :
Lecture :
1 - Introduction to the course, luminance images, color images + history of photography and images. (O. Strauss)
2 - Image sensors in the modern world. (P. Falgayrettes)
3 - Non-photonic image acquisition technology: ultrasound images, microscopic images, radar images, mechanical images, thermal images, etc. (P. Falgayrettes)
4 - Color representation in images, perceptual colors. Reconstructed images (tomography). (O. Strauss)
5 - Histograms and image binarization. (O. Strauss)
6 - Morphology, basic erosion/dilation operators. Extension to grayscale morphology (O. Strauss)
7 - The image as a sampled and quantized signal. (P. Falgayrettes)
8 - Grayscale image filtering, principle and convolution masks. (P. Falgayrettes)
9 - Image denoising, high-pass, low-pass, band-pass filters. (P. Falgayrettes)
10 - Interpolation kernels, convolution kernels. (O. Strauss)
11 - Deconvolution, image derivation. (O. Strauss)
12 - Contour extraction, extraction of particular points (Harris). (O. Strauss)
13 - Chaining contours, eliminating holes ... (P. Falgayrettes)
14&15 - Fourier transform on images. (P. Falgayrettes) (3h00)
16 - Physical interpretation of images (back to the model). (B . Wattrisse)
17 - Geometric transformations on images. (O. Strauss)
18 - Correlations and correlation-related methods. (O. Strauss)
19 - Object parameterization, geometric invariants, inertia matrices, etc. (P. Falgayrettes)
20 - Principles of compression. (P. Falgayrettes)
21 - Segmentation of images into regions. (O. Strauss)
Practical work
1 - Basics of image processing, binarization, morphology, sampling, quantization.
2 - Color representation: physical space, perceptual space.
3 - Fourier transform on images.
4 - Image filtering, discrete convolution.
5 - Sampling, interpolation, geometric transformation.
6 - Image derivation, contour extraction, Harris points.
Contact
Manager: O. Strauss - LIRMM, UM
Administrative contact(s): Claudie Fabry