Deep Learning for Image Processing: From the theory to the Industry 4.0

What you will learn

  • Understand how a neural network learns and what it needs to learn well.
  • Tune your neural network and use the right techniques to solve common problems that occur in real industrial applications to ensure better performance.
  • Convolutional Neural Networks CNN and how they are used for image classification, object detection in images, and image segmentation.
  • Several industrial use cases and examples will be presented all along the course for a better understanding of real applications.

Syllabus

Chapter 1: Introduction to Machine Learning

  • What is Machine Learning?
  • Machine Learning and statistics
  • Types of Machine Learning models

Chapter 2: Introduction to Neural Networks

  • Differences between Machine Learning and Deep Learning
  • What is a Neural Network? and how does it learn?

Chapter 3: Convolutional Neural Networks

  • Introduction to convolutional neural networks
  • Structure and advantages of a convolutional neural network

Chapter 4: Training a neural network

  • Preparing the dataset and its annotation
  • Convergence and overfitting: Is my network learning well?

Chapter 5: Training tips and tricks

  • Training techniques: Common problems during the training and how to solve them

Chapter 6: Convolutional Neural Network for Object Detection

  • From Classification to Object Detection: an overview

Chapter 7: Convolutional Neural Network for Segmentation

  • From Classification to Segmentation: an overview

Prerequisites

  • Basic knowledge in mathematics and algebra
  • Basic level in English
  • A PC and notebook

Target learners

Anyone wishing to acquire a specialization as a Deep Learning Engineer for image processing

Duration

32 hours in 2 months

Teaching mode: Hybrid

Start date: Saturday, 03 December 2022

End date: Saturday, 28 January 2023

Professor in charge

Prof. Karim Tout

Senior Computer Vision / Machine Learning Engineer at Uqudo. Karim received an engineering degree and a master degree in electrical and system control in 2014, followed by a PhD degree in Computer Vision in 2018. Karim worked for four years as an Industrial Postdoctoral Researcher on Computer Vision and Machine/Deep Learning projects. After that, Karim occupied the position of Lead Machine Learning Engineer at Cetim Grand Est managing a team of data engineers before joining Uqudo in 2022. Karim specializes in Computer Vision, Machine Learning, Deep Learning and Image Processing applied to industrial applications.

en_USEnglish