Cours : MAI5021 - Image Analysis and Representation (2023) | e-Disciplinas

Aperçu des sections

Généralités

Tout replier Tout déplier
- Announcements and News Forum
- Git Repository with the Notebooks used on lectures URL
- Intensive course. Period: 20/03 - 08/05
  Course codes:
  Undergrad: SCC0251, full course compressed to 8 weeks.
  Post-grad: MAI5020, first half of course in the same 8 weeks
  Post-grad: MAI5021, second half of course in the same 8 weeks. This is the edisciplinas you are seeing
  
  Support Team:
  The support team will design the assignments and be available throughout the course for questions regarding tests, assignments and classes. Questions should be sent through the Discord platform:
  Discord Server Invite Link: https://discord.gg/4ZBeVpQTqz
  DM to Leo on Discord: a_leo#4871
  DM to Gustavo on Discord: GustavoAraujo#5994
- Course Presentation Fichier
- [Video] Course Presentation URL
  
  Non disponible à moins que : Vous soyez membre d’un groupe
Fundamentals of Image Processing + Color Image Processing
- Target period: 20/03 to 26/03
  There are two exams this week, the Exam - Fundamentals and the Exam - Color Image Processing
  Fundamentals
  Class information and Introduction to image processing
  Digital image processing fundamentals
  Image acquisition, quantization, sampling
  
  Color Image Processing
  Color and light spectrum
  Color models: RGB, CMYK, XYZ, Lab, HSV
  Color image processing
- DIP_01_Fundamentals Fichier
- [interactive video 1-1] Image, digital image and sensors Contenu interactif
  
  Terminer l’activité
- [interactive video 1-2] Natural vision and quantization Contenu interactif
  
  Terminer l’activité
- [interactive video 1-3] Color quantization and brightness perception Contenu interactif
  
  Terminer l’activité
- [video 1-4] Coding environment Python 3 and Jupyter Notebook URL
  
  Consulter
- [interactive video 1-5] Implementing an image processing program in Python Contenu interactif
  
  Terminer l’activité
- DIP_06_Color Fichier
- [interactive video 6-1] Introduction: what is color, light spectrum Contenu interactif
  
  Terminer l’activité
- [interactive video 6-2] Color models: RGB, CMYK, XYZ, Lab, HSV Contenu interactif
  
  Terminer l’activité
- [interactive video 6-3] Examples of color image processing in Python Contenu interactif
  
  Terminer l’activité
- Exam - Fundamentals of Image Processing Test
  
  Recevoir une note
- Exam - Color Image Processing Test
  
  Recevoir une note
Image Description and Representation
- Target period - 27/03 to 03/04
  Color and texture descriptors
  Texture: Haralick and LBP
  Bag of Features
- DIP_07_TextureAnalysis Fichier
- [interactive video 7-1] Basic color and texture descriptors Contenu interactif
  
  Terminer l’activité
- [interactive video 7-2] Color histograms and distance functions in python Contenu interactif
  
  Terminer l’activité
- [interactive video 7-3] Texture descriptors: Haralick (GLCM) and LBP Contenu interactif
  
  Terminer l’activité
- [video 7-4] Texture analysis with LBP implemenation in python and skimage URL
  
  Consulter
- [interactive video 7-5] Bag of Features Contenu interactif
  
  Terminer l’activité
- Exam - Image Description and Representation Test
  
  Consulter
Mathematical Morphology
- Target period - 03/04 to 17/04
  Structuring elements
  Dilation and erosion
  Opening and closing
  Graylevel image processing with Morphology
- DIP_08_Morphological_processing Fichier
- [interactive video 8-1] Mathematical Morphology definitions and basic operators URL
  
  Consulter
- [interactive video 8-2] Dilation and Erosion in Python URL
  
  Consulter
- [interactive video 8-3] building new operators: opening and closing URL
- [interactive video 8-4] Implementing the hit-or-miss in python URL
  
  Consulter
- [interactive video 8-5] Graylevel image processing with morphology URL
  
  Consulter
- Exam - Mathematical Morphology Test
Image Segmentation
- Target period - 10/04 to 24/04
  Thresholding-based segmentation
  Otsu method
  Edge detection and region-based segmentation
  Watershed segmentation
  Hough transform
- DIP_09_Segmentation Fichier
- [interactive video 9-1] Segmentation and threshold-based methods Contenu interactif
  
  Terminer l’activité
- [interactive video 9-2] Thresholding and Otsu in Python Contenu interactif
  
  Terminer l’activité
- [interactive video 9-3] Edge detection and Region-based segmentation Contenu interactif
  
  Terminer l’activité
- [video 9-4] Region-based pixel aggregation in python URL
  
  Consulter
- [interactive video 9-5] Introduction to Hough transform and (very brief introduction) to Watershed segmentation Contenu interactif
  
  Terminer l’activité
- [interactive video 9-6] Watershed implementation in Python and Distance Transform Contenu interactif
  
  Terminer l’activité
- Exam - Image Segmentation Test
Neural Networks for Images
- Target period - 17/04 to - 30/04 (Two Weeks)
  Deep learning
  Image classification
  Convolutional networks
  CNN architectures
- DIP_10_CNNs Fichier
- [interactive video 10-1] Deep Neural Networks for Image Classification - Motivation Contenu interactif
  
  Terminer l’activité
- [video 10-2] Machine Learning vs Deep Learning URL
  
  Consulter
- [video 10-3] Optimizing/learning a linear classifier URL
  
  Consulter
- [interactive video 10-4] Dense Network for Digit Recognition Contenu interactif
  
  Terminer l’activité
- [interactive video 10-5] Convolutional Networks Contenu interactif
  
  Terminer l’activité
- [video 10-6] Popular CNN Architectures URL
  
  Consulter
- [interactive video 10-7] Training Strategies and Limitations Contenu interactif
  
  Terminer l’activité
- Exam - Neural Networks for Image Classification Test
Assignment 01: image descriptors
- Opening date: 27/03
  Deadline: ~~09/04~~ 11/04
  In this assignment you have to implement a classic feature extractor called Histogram of Oriented Gradients (HoG) and use it to train a machine learning model for people detection.
  Remember to check your code with run-codes-local. The grade given by run-codes-local will compose your final grade for this assignment.
- Specification for Assignment 01 Fichier
- Test Cases for Assignment 01 Fichier
- Submission for Assignment 01 Devoir
Assignment 02: morphology
- Opening date: 10/04
  Deadline: 24/04
- Specification for Assignment 02 Fichier
- Test Cases for Assignment 02 Fichier
- Submission for Assignment 02 Devoir