Spatial Analysis, 3D Data & Machine Learning

Coursera via Coursera

Go to class Write review

Details

Go to class

Provider

Coursera
Pricing

Paid Course
Languages

English
Certificate

Certificate Available
Effort

10 hours
Sessions

Self-Paced
Level

Beginner
Subtitles

English

Found in

Part of

Mastering Geospatial Data Science: From Beginner to Expert

Overview

AI, Data Science & Cloud Certificates from Google, IBM & Meta — 50% Off

One plan covers every Professional Certificate on Coursera. 50% off Coursera Plus Annual for 10 days only — price increases June 17.

Unlock All Certificates

Advance your skills in spatial analysis and machine learning for geospatial data. This course covers geostatistics, LiDAR and 3D data processing, and supervised machine learning techniques. You will also learn how to apply deep learning methods for imagery analysis. By the end of the course, you will be able to build and evaluate models for geospatial data and analyze complex spatial patterns.

Syllabus

Crunch Spatial Stats: Detecting Spatial Patterns with Global Moran’s I

In this module, you will explore how spatial patterns differ from random distributions and why that difference matters in real-world analysis. Using air-quality sensor data as a motivating example, you will examine how Global Moran’s I quantifies spatial autocorrelation in polygon data and helps analysts identify clustering patterns that might otherwise go unnoticed.

Crunch Spatial Stats: Estimating Surfaces with IDW Interpolation

In this module, you will examine how spatial analysts estimate values between discrete measurement locations. Using air-quality sensor data as a motivating example, you will be introduced to Inverse Distance Weighting (IDW) interpolation and learn how distance-based assumptions are used to generate continuous surfaces from point observations. You will explore how parameter choices influence interpolation results and learn how to interpret estimated surfaces responsibly in real-world spatial analysis contexts.

Crunch Spatial Stats: Understanding Spatial Autocorrelation with Semivariograms

In this module, you will step back from computation to interpretation, focusing on semivariograms as diagnostic tools for spatial structure. By learning how to read range, sill, and nugget, you will gain intuition about spatial dependence, knowledge that informs both analysis choices and communication with non-technical audiences.

Explore LiDAR in 3D: Visualize LiDAR Point Clouds in 3D

Learners understand what LiDAR point clouds represent and can confidently load and explore them in a 3D environment.

Explore LiDAR in 3D: Generate a DEM from Ground-Class Points

Learners understand why DEMs are derived products and can create one correctly from ground-class LiDAR points.

Explore LiDAR in 3D: Assess Vertical Accuracy Using Control Points

Learners evaluate whether a DEM is fit for purpose by comparing it against known reference elevations.

Train ML Models: From Pixels to Predictors

You will explore why raw imagery alone is insufficient for supervised classification and how engineered features improve model performance. The lesson focuses on practical extraction of spectral bands and texture metrics used in land-cover analysis.

Train ML Models: Training a Random Forest Classifier on Imagery Data

You will apply engineered features to train a Random Forest classifier. Emphasis is placed on intuition: how trees vote, how parameters affect performance, and how to avoid beginner mistakes.

Train ML Models: Evaluating Accuracy: Confusion Matrices & Model Validation

You will evaluate whether the model meets job requirements by interpreting confusion matrices and accuracy metrics. The lesson emphasizes decision-making, not just calculation.

Deep Learn Imagery: Fine-Tuning CNNs for Land Cover

In this module, you will apply transfer learning techniques to fine-tune a pre-trained convolutional neural network (CNN) for land cover classification using satellite imagery. The module focuses on adapting existing vision models to geospatial data under real-world constraints such as limited labeled samples, class imbalance, and spatial generalization challenges.

Deep Learn Imagery: Improving Model Performance with Data Augmentation

In this module, learners design and apply data augmentation pipelines to improve the generalization of convolutional neural networks trained on satellite imagery. The module focuses on selecting realistic augmentations that preserve spatial meaning while addressing limited and imbalanced land-cover data.

Deep Learn Imagery: Explaining Model Predictions with Grad-CAM

In this module, learners use Grad-CAM visualizations to interpret convolutional neural network predictions for satellite imagery. The module emphasizes understanding model attention, identifying failure modes, and communicating model behavior clearly to technical and non-technical stakeholders.

Project Module: Geospatial Machine Learning

In this project, you will build a geospatial machine learning workflow to classify land cover using imagery, LiDAR-derived elevation data, and labeled samples. You will engineer features, train a model, validate the results, and generate a classified land cover output. You will also summarize model performance and create an interpretation output to explain how the model behaves. This project requires learners to demonstrate spatial analysis, 3D data use, machine learning implementation, validation, interpretation, and stakeholder communication in one authentic workflow.