Manipulation of digital images
From enhancement to interpretation
During this course, learners will discover the answers to the following questions:
- What techniques are used to filter images and improve their representation?
- How do we go from analysis to computer vision?
- Can we develop robotic systems with vision?
The course lasts 6 days (36 hours).
In 2 parts of 3 days, it includes many exercises and demonstrations.
The digital image is becoming more and more present and software that accompanies digital devices is increasingly sophisticated.
This course offers a simple approach to the processing of digital images from examples. It introduces some of the basic principles of sampling and filtering and gives an overview of the algorithmic techniques used both in the field of research and in an industrial setting.
Course level: Basic/Advanced
Knowledge of basic math functions (logarithm, exponential, linear combination).
Basic knowledge of statistics (mean, standard deviation, median).
Associate Professor of Applied Physics, Doctor of the University of Orsay - Expert in Image Processing, Information System Project Manager at the CGLLS.
- Principles of digital imaging
- Image acquisition (sampling, quantification, color space), visualization
- Image enhancement: algorithmic techniques to enhance digital images
- Noise, scanning artifacts and associated filters
- Contrast enhancement and restoration
- Examples of grayscale (visible and infrared) image enhancement and color enhancement
- Current trends in image processing
- Implementation of algorithms on graphics cards (GPGPU)
- Measure less, measure blur, rebuild better: how signal processing is changing the design of optical systems
- Deep Learning
- Advanced filtering techniques
- Advanced filtering: introduction to mathematical morphology for pattern detection and counting
- Application to particle size
- Texture, introduction to the analysis of natural images
- Introduction to the use of wavelets in image processing
- Applications: detection, recognition, identification
- Introduction to classification for pattern recognition: descriptors and classifiers, segmentation techniques (statistics, geometry, optimization)
- Detection and recognition: application of segmentation and classification; which descriptors for which techniques?
- Assembly of different treatments: analysis, filtering, detection and recognition
- Use of image processing in some concrete cases
- Exercises in computer room: images, problems to solve with the tools seen/reflection then correction. Using a program library on Raspberry PI-2
Scheduled in French:
PARIS: Part 1: 10 to 12 May 2021
PARIS: Part 2: 25 to 27 May 2021
For the English realization, please, consult us.
€2,460 excluding tax (20% VAT)