Land Cover Classification in Google Earth Engine

Name: Land Cover Classification in Google Earth Engine
Rating: 4.2 (6 reviews)

Start applying machine learning for remote sensing

Created byTarık Emre Yorulmaz

Last updated 5/2025

English

What you'll learn

Get Theoretical Knowledge of Random Forest Algorithm
Proficiency in Google Earth Engine
Training Data Development
Land Cover Mapping
Accuracy Assessment

Course content

4 sections • 18 lectures • 1h 9m total length

What is Google Earth Engine?1:35
Explore geospatial analysis with Google Earth Engine, a cloud-based platform for satellite imagery. Study environmental datasets and trends from three decades, including deforestation, urban expansion, water resources, and land use.
Google Earth Engine Interface1:46
Explore the Google Earth Engine interface by clicking Get Started, selecting a Gmail account, registering a project, and opening the code editor; commercial use is paid, but research is free.
Landsat 8 Image Visualization17:30

What is Random Forest?2:10
Apply random forest to land cover classification in Google Earth Engine by training with labeled Landsat 8 pixels, using many decision trees and majority voting on spectral bands and NDVI.
Overfitting2:25
Overfitting in land cover classification on Landsat data occurs when a random forest memorizes pixels; prevent with bagging, random features, diverse data, balanced classes, limited depth, many trees, and cross-validation.

Create Training Dataset6:29
Combine Training Points4:21
Choose Features3:37
Get Feature Values3:33
Collect red and IR feature values, plus ndvi, at each training point to build a dataset for random forest learning on Landsat imagery.
Split Data for Training and Testing4:03
Split a 113-point land cover dataset into 70% training and 30% testing for a random forest in Google Earth Engine, and verify the sizes in the console to prevent overfitting.
Train Random Forest2:48
Train a 50-tree random forest in Google Earth Engine from 80 training points to classify pixels as water, vegetation, urban, or bare land using four features.
Classify the Image1:35
Apply a trained random forest to label every pixel in Qmra as water, vegetation, urban, or bare land, producing a land cover map in Google Earth Engine.
Land Cover Mapping4:38
Map land cover in google earth engine with blue water, green vegetation, grey urban, and yellow bare land, using random forest with training points and red, nir, swir1, ndvi features.
Potential Questions and Answers1:23
Explore why four features balance speed and accuracy in land cover classification with random forest, why data splitting matters, and how 53 and testing points guide performance without overfitting.

Intro0:53
Classify the Validation Set1:32
Use the trained random forest to predict the class of each testing point (water, vegetation, urban or Maryland) and compare these predictions to the true labels from video four.
Create a Confusion Matrix7:52
Potential Questions and Answers1:28
Use a confusion matrix to reveal misclassified areas, like water vs vegetation, and adjust training points and features to improve producer's and user's accuracy.

Requirements

A free Google Earth Engine account (enrollment instructions provided)
Access to a computer with a reliable internet connection

Description

Welcome to an in-depth and rigorously structured course designed to equip learners with the expertise to perform land cover classification using Random Forest within Google Earth Engine (GEE). This course is tailored for students, geospatial professionals, environmental scientists, and researchers seeking to harness satellite imagery for precise land cover mapping. Through a comprehensive case study in Çumra District, Konya, Türkiye, participants will develop proficiency in classifying land into four categories—Water, Vegetation, Urban, and Bare Land—utilizing state-of-the-art machine learning techniques and cloud-based geospatial platforms. No prior experience in coding or remote sensing is required, as this course provides a systematic progression from foundational concepts to advanced applications, ensuring accessibility for beginners and value for experienced learners.

Upon completion, you will produce a professional-grade land cover map of Çumra District, demonstrating mastery of Random Forest and GEE. You will gain the ability to preprocess satellite imagery, develop and validate machine learning models, and interpret geospatial data, skills highly valued in academia and industries such as environmental management, urban planning, and agricultural monitoring.

Embark on a transformative learning journey to master land cover classification with Random Forest in Google Earth Engine. This course offers a unique opportunity to develop cutting-edge skills through a practical, real-world project in Çumra District, equipping you to address global environmental challenges. Enroll now to gain expertise in geospatial analysis, contribute to sustainable development. Begin your journey today and unlock the potential of satellite imagery to map and understand our world.

Who this course is for:

Undergraduate and graduate students in environmental science, geography, or related fields seeking practical geospatial skills
Geospatial professionals aiming to integrate machine learning and GEE into their workflows.
Researchers and analysts interested in leveraging satellite imagery for environmental and urban studies

Land Cover Classification in Google Earth Engine

What you'll learn

Explore related topics

Course content

Introduction3 lectures • 21min

Random Forest Algorithm2 lectures • 5min

Land Cover Classification9 lectures • 32min

Accuracy Assesment4 lectures • 12min

Requirements

Description

Who this course is for: