ials / PR2023

Percepción Remota

Maestría en Geomatica - Código curso 2020039 - 2023-1S

This is the main communication site for the course. Here you will find the lectures, readings, and assignments. This is a place to visit regularly.

Objectives

The purpose of this course is to strengthen the theoretical foundations of remote sensing and to develop the student's skills for digital image processing and analysis. This course focuses on developing critical thinking to conduct research using earth observation data.

The course syllabus is available at this link

SCHEDULE:

PART 1 - FUNDAMENTALS

Week 1 – Remote sensing and digital image processing – Cap. 1 from Jensen (2015)

Topics:

• ¿What is remote sensing?
• ¿What is a remote sensing system?
• ¿What are advantages and limitations of remote sensing?
• ¿What is the remote sensing process?
• ¿What are the key characteristics of digital images?

In-class exercises:

• Search and download satellite images
• Visualize multispectral images

Independent work:

• Module 1: Introduction to Google Earth Engine

Week 2 - Electromagnetic radiation – Cap. 2 from Campbell (2022)

Topics:

• The Electromagnetic Spectrum
• Major Divisions of the Electromagnetic Spectrum
• Radiation Laws
• Interactions with the Atmosphere
• Interactions with Surfaces

In-class exercises:

• TBW

Independent work:

• Module 2: Exploring GEE Datasets

Week 3 – Remote sensing of vegetation - Cap. 11 from Jensen (2014)

Topics:

• Leaves
• Plants
• Canopies

In-class exercises:

• Start Module 3 below

Independent work:

• Replicate these tutorials with data for your study area:
  • Module 3: Raster interpretation and visualization
  • Module 4: Moving Data In and Out of Google Earth Engine
    • Download IDEAM landcover data from here

REFERENCIA

• Cómo usar Google Earth Engine y no Fallar en el Intento

PART 2 – IMAGE ACQUISITION

Week 4 – The real potential of remote sensing to map aboveground biomass in tropical forests (Nidhi, 2022)

Topics:

• Presentation of Nidhi's article
  • P
• Discussion D

In-class exercises:

• Start module 5 below

Independent work:

• Replicate these tutorials with data for your study area:
  • Module 5: Raster Pre-processing

Week 5 – Image interpretation – Cap. 6 from Campbell (2022)

Topics:

• What is image interpretation?
• Why image interpretation is useful?
• Image Interpretation Tasks
• Elements of Image Interpretation
• Classification schemas

In-class exercises:

• Start module 6 below

Independent work:

• Replicate this tutorial with data for your study area:
  • Module 6: Developing Indices in Google Earth Engine
• Create "fichas técnicas" for the land cover classes you are interested in:
  • Reference

Week 6 – Advances and trends in remote sensing (Ustin & Middleton, 2021)

Topics

• Spectral indices for detecting built-up land areas

In-class exercises:

• Follow these tutorials with data for your study area:
  • Plotting spectral response curves
  • Spectral signatures

Independent work:

• Replicate this tutorial with data for your study area:
  • Module 7: Image classification

PART 3 – IMAGE PROCESSING

Week 7 – Atmospheric correction

Topics

• Remote sensing atmospheric correction
• Bayesian atmospheric correction over land

In-class exercises:

• Follow this tutorial with data for your study area:
  • SIAC on GEE

Students will submit the first report before 22th March at 12:00 midnight (25% of the final grade)

Independent work:

• Do the second iteration of your random forest (RF) land cover classification in GEE
• Make sure you document each iteration

Week 8 – Modern photogrammetry (Forsmoo et al., 2019)

Topics:

• Essential reference:

  • 3D Structure from 2D Motion

• Additional references:

  • Epipolar geometry
  • Stereo systems

Students enrolled in the course will write the first examination on 29th March (25% of the final grade)

• List of enrolled students

• Make sure you register in the canvas platform before coming to the classroom

• Independent work:

  • Time series analysis in GEE

Week 9 – Geometric corrections

• Topics

  • Structure from Motion
  • Pre-processing UAV images

• Reference

  • Processing Coastal Imagery With Metashape

• Independent work:

  • TBD

• Notas Informe No. 1

PART 4 – IMAGE ANALYSIS

Week 10 – Basics of machine learning (ML)

See the recording of this week lecture here

Topics:

• Basics of ML:

  • What is ML?
  • What are decision trees (DT)?
  • What are random forests (RF)?

• Sample size for accuracy assessment

• In-class exercise:

  • Generating your own samples

• Independent work

  • Adapt the sampling generation exercise to your image analysis problem

Week 11 - Machine learning of ecological variables

Topics:

• ML-based maps of ecological variables

• Predicting into unknown space

• In-class exercise:

  • RF Species Classification

• Independent work

  • In case it is useful, adapt the GEE RF code to analyze your own images

Week 12 - Object-based image analysis (OBIA)

Topics:

• OBIA - Overview

Reference:

• GEOBIA – Towards a new paradigm

• In-class exercise:

  • GEOBIA in GEE
  • Object-based methods in GEE

• Independent work

  • Continue iterating & writing your Informe No. 2

Week 13 - Texture metrics

• Topics:

  • GLCM textural metrics for remote sensing image analysis

• In-class exercise:

  • Neighborhood based Image Transformations in GEE

• Readings:

  • The Comparison of Different Methods of Texture Analysis for Their Efficacy for Land Use Classification in Satellite Imagery
  • Practical guidelines for choosing GLCM textures to use in landscape classification tasks over a range of moderate spatial scales

• Independent work:

  • Include textural metrics in your covariables dataset and examines its effects
  • Continue iterating & writing your Informe No. 2

Week 14 - Machine Learning for Remote Sensing

• Topics:

  • There is no face-to-face class.
  • Instead, make sure you attend the following virtual sessions from NASA (and take notes):
    • Part 1: Overview of Machine Learning
    • Part 2: Training Data and Land Cover Classification Example

• Independent work:

  • Complete & revise your Informe No. 2
  • Plazo para entrega de Informe No. 2
    • 24 de Mayo - 2:00 pm -
    • Impresión a color

• Notas Examen No. 1

Week 15 - Machine Learning for Remote Sensing (cont.)

• Topics:

  • There is no face-to-face class.
  • Instead, make sure you attend the following virtual session from NASA (and take notes):
    • Part 3: Model Tuning, Parameter Optimization, and other ML Algorithms

• Independent work:

  • Complete & revise your Informe No. 2

  • Nuevo plazo para entrega de Informe No. 2

    • 26 de Mayo - dejarlo en Oficina 333 antes de las 11:59 am -
    • Impresión a color
    • Sobre sellado marcado con su nombre

  • Study for Examen No. 2 (to be written on 31 May at 2:00 pm)

Week 16 - NOTAS

• Notas - 07.06.2023
• SIA - 12.06.2023