Virtual Classroom

Object Detection - Remote training

DETECOB-EN-D

Who should attend?

  • Deep learning, a branch of artificial intelligence has recently provided some new approach to recognize objects (patterns) in images and videos. In the domain of Oil & Gas activities, this has a lot of applications like:
  • seismic interpretation,
  • seismic attributes,
  • micro-paleontology,
  • petrography,
  • video surveillance.
  • The object detection has proven his efficiency in some domain to increase the O&G engineers capabilities, this training is about to explain by lecture and practical exercises the fundamentals of theses algorithms.
Public :
  • This training has been designed for geoscientists who have to work with data scientists to develop object detection tools, or for geoscientists who have to use object detection tools to reach their business objective. The notebook workshops are designed for people having or not a python experience, but some general knowledge of coding is required.

Level :Awareness

Prerequisite :
  • No pre-requisistes are necessary to follow this course.

Course Content

    Day 1
  • INTRODUCTION

      • Problem statement.
      • Machine learning versus deep learning.
  • CLASSIFICATION USING A CNN ALGORITHM

      • Neural network and CNN.
      • Classifying hand written numbers using a CNN:
      • MNIST dataset.
      • Building a LeNet5 notebook.
      • Analyzing the benchmark, how to measure the performance of a model.
      • What does happen is the image dataset is more complex: classifying CIFAR-10 dataset.
    Day 2
  • THE R-CNN FAMILY ALGORITHMS

      • R-CNN.
      • Fast R-CNN.
      • Faster R-CNN.
  • THE YOLO APPROACH

  • UNBOXING YOLO, TESTING IT ON A VIDEO

      • Training a YOLO architecture.
      • Performance.
    Day 3
  • IMPROVING IMAGES WITH GAN - GENERATIVE ADVERSARIAL NETWORKS

      • How GAN works, discriminative and generative models.
      • Generator and discriminator models.
      • Improving object detection with GAN.
      • Application in seismic modeling.
  • FROM UNSUPERVISED TO SUPERVISED MODELS FOR SEISMIC INTERPRETATION

      • The issue with supervised CNN.
      • 3 unsupervised classification of seismic data:
      • K-means clustering.
      • Agglomerative hierarchical clustering.
      • Kohonen self-organizing feature map (SOFM).
      • Supervised models for seismic interpretation, the state of the art.

Learning Objectives

  • At the end of the training, the trainees will be able to:
  • Know about the various strategies offered by deep learning to detect objects,
  • Measure the accuracy of the models,
  • Understand the current limitation of object detection.

Ways & Means

  • This course can be delivered using in-person lectures or virtual classrooms. Each training module contains lectures, hands-on practices and/or case studies.

More

Coordinator :IFP Training trainers (permanent or contracted) having a good expertise and/or experience of the related topics, trained to adult teaching methods, and whose competencies are kept up-to-date.

To French entities : IFP Training is referenced to DataDock ; you may contact your OPCO about potential funding. Please contact our disabled persons referent to check the accessibility of this training program : referent.handicap@ifptraining.com