Image processing in remote sensing (2022) - ENSG - lectures and practical lessons

In this course, we introduce the basics of remote sensing image analysis techniques. The module can be divided into several parts: we start by explaining the basics of remote sensing data acquisition and analysis (difference between optical and radar images, overview of the tasks and corresponding solutions, ...), then we introduce the supervised and non-supervised learning algorithms for land cover mapping (kNN, Random Forest, K-means, ...), finally, we present the concept of object-based image analysis and associated texture extraction.

All the practical lessons were realized in Orfeo ToolBox and QGIS. Some Python coding was used for the automatisation.

Machine Learning and Deep Learning for Remote Sensing Data Analysis (2021-now) - ENSG - practical lessons

I gave multiple practical lessons for data analysis in remote sensing using Python. Those courses covered large spectrum of the algorithms including the traditional machine learning algorithms, such as Random Forest and SVM for pixel-wise image classification, as well as deep learning algorithms for object classification, and semantic segmentation of optical images, as well as 3D point cloud data.

Data Analysis (2018-2020) - ISEP - practical lessons and development of teaching supports

In this course, we introduce several key tools from data analysis: This course first covers some basic elements of statistic that are useful to analyse numerical data (univariate, bivaritate and multivariate statistics, estimates, and confidences intervals). It also explains how to deal with different types of data : numerical, binary, categorical, text data and time series, all the while providing specific tools for each type of data. This course also gives some important elements of data visualization including techniques such as PCA, ISOMAP, LLE and t-SNE that are useful to visualize high dimensional data. And finally, 2 classes are given about clustering and classification and can be seen as a brief introduction to Machine Learning and its most basic methods and concepts.

The first year R language was used as the programming support, the second year, all the exercises were translated to Python as the more adapted language for the real-life problems.

Java and Algorithms (2017-2020) - ISEP - practical lessons

This course is an introduction to different concepts of algorithmic with Java language as support. Different courses offers a progressive approach of algorithmic starting by the methodology: understanding the problem, splitting into tasks, writing a pseudo-code and reflection on the using structures.

The Java language and the basic of UML are introduced along with the methodology with multiples examples of solving the classic algorithms (sorting algorithms, lists and stacks manipulations, etc).