DESCRIPTION

Nowadays, digital images and video are everywhere. Image Processing revolutionizes very many domains, notably:

  1. Digital Media (video/image/movie) Content Production and Broadcasting, Social Media Analytics,
  2. Medical/Biological/Dental Imaging and Diagnosis,
  3. Big Visual Data Analytics,
  4. Internet and Communications (media broadcasting, streaming).
  5. Scientific Imaging of any sort, e.g., Remote Sensing, Environment Sensing.

Photoshop and many other image processing tools are ubiquitous.

Furthermore, Image Processing  is typically the first step that enables diverse applications, in unison with Computer Vision and Machine Learning:

  1. Autonomous Systems (cars, drones, vessels) Perception,
  2. Robotics Perception and Control,
  3. Intelligent Human-Machine Interaction,
  4. Anthropocentric (human-centered)Computing,
  5. Smart Cities/Buildings and Assisted living.

Visual Computing, encompassing Computer Vision and Image Processing, coupled with AI (notably Machine Learning and Deep Neural Network) advances hit the news almost every day.

This CVML Web Module focuses on Image Processing and 2D Signal Processing theory, its applications in the above-mentioned diverse domains and new challenges ahead.  First, an Introduction to Image Processing will be offered to clarify concepts in a precise and mathematical way, to be complemented by a formal Image Typology. Image sampling will provide the necessary background to understand the potential and limitations of digital images. It will be complemented by a mathematical and programming definition and treatment of digital images. Image formation and its issues (e.g., image noise, deformations) will then be detailed, whether based on visible light or on other modalities (e.g., Xrays, Ultrasound). 

2D Signals and Systems will provide the theoretical and algorithmic tools for most image processing operations. The basic 2D Signal Processing  will be detailed in 2D Digital Filter Design and Realization. Fast 2D convolution algorithms will be presented in detail, as they will provide efficient implementation of most image processing operations and are the cornerstone of Convolutional Neural Networks and of many Computer Vision Tasks. Then notions related to Image transforms will be clarified, together with their applications in image/video analysis and compression. 

All the above will be used to model precisely and mathematically Digital Image Formation and Digital Camera structure. Human Visual System will be overviewed. It provides the foundations for  Image Perception. Particular attention will be paid to Color Theory.  They both have a strong impact on Image Quality, Computational Aesthetics and Image Processing system design specifications. Image filtering will provide tools to reduce noise and enhance image quality, e.g., to increase contrast, perform image zooming or printing.  Special image processing operations, will be also overviewed, notably halftoning, pseudo-coloring, contrast enhancement, Super-resolution, High-Dynamic Range Imaging and Image Restoration.

As digital images are ubiquitous and have large memory demands, Image Compression will be presented in detail.

LECTURE LIST

  1. Introduction to Image Processing
  2. Image Typology
  3. Image Sampling
  4. Digital Images
  5. 2D Systems
  6. 2D Digital Filter Design and Implementation
  7. Fast 2D convolution algorithms
  8. Image Transforms
  9. Digital Image Formation
  10. Human Visual System
  11. Image Perception
  12. Color Theory
  13. Image Quality
  14. Computational Aesthetics
  15. Digital Image Filtering
  16. Digital Image Processing
  17. Super-resolution
  18. High-Dynamic Range Imaging
  19. Digital Image Restoration
  20. Digital Image Compression
  21. Neural Image Compression
CVML WEB LECTURE MODULE SCHEDULE

This module has been designed to be mastered within 1 month (or less), if you have proper background (at least early undergraduate student in an EE, ECE, CS, CSE or any Engineering or Exact Sciences Department).
We propose that you follow the above mentioned  Lecture order. You may want to skip few Lectures that might not be of immediate interest to you for later study.

On average you can study 4 lectures per week. The related effort is as follows:
1) Lecture pdf study and filling the related understanding questionnaire: 1-2 hours per lecture (on average, depending on your background)
2) Tutorial exercise (if available): 1/2 hour on average (more if you do not have theoretical skills). We strongly recommend to try solve them yourself, before resorting to the existing solution.
3) Programming exercise (if available): 3-4 hours on average (more if you do not have good programming skills). We strongly recommend to try program them yourself, before resorting to the existing code.

The following lectures are accompanied by programming or tutorial exercises:

  1. Image Sampling (2 Tutorial Exercises)
  2. Digital Images (1 Programming Exercise)
  3. 2D Systems (7 Tutorial Exercises)
  4. 2D Digital Filter Design and Implementation (2 Tutorial Exercises)
  5. Fast 2D Convolution Algorithms (1 Programming Exercise)
  6. Image Transforms (9 Tutorial and 1 Programming Exercises)
  7. Digital Image Formation (1 Tutorial Exercise)
  8. Image Perception (2 Tutorial Exercises)
  9. Color Theory (1 Tutorial and 1 Programming Exercises)
  10. Digital Image Filtering (2 Tutorial and 15 Programming Exercises)
  11. Digital Image Processing (3 Tutorial and 5 Programming Exercises)
  12. Digital Image Compression (4 Tutorial Exercises)