Cloud-Based Remote Sensing with Google Earth Engine

Table of Contents

1. Frontmatter

   Download PDF-version

2. Programming and Remote Sensing Basics

   1. Frontmatter

   2. Chapter 1. JavaScript and the Earth Engine API

      • Open Access

      Ujaval Gandhi

      This chapter provides an introduction to the Google Earth Engine API and JavaScript syntax, essential for scientific data analysis. It guides users through the Code Editor environment, teaching basic programming concepts such as variables, data structures, and functions. The chapter emphasizes the practical use of the Earth Engine API for efficient computations and the integration of cloud-hosted datasets. By the end of the chapter, readers will be familiar with the Earth Engine platform and ready to start building scripts for remote sensing analysis.

      AI Generated

      This summary of the content was generated with the help of AI.

      Show original text

      Download PDF-version

   3. Chapter 2. Exploring Images

      • Open Access

      Jeff Howarth

      This chapter delves into the power of satellite images in Google Earth Engine, focusing on visualizing and interpreting data stored in image bands. It begins by teaching readers how to load and visualize individual bands as map layers, then progresses to creating composite layers that combine multiple bands. The chapter compares true-color composites, which resemble natural colors, with false-color composites that reveal information outside the visible spectrum. It also explores the additive color system and its application to more abstract attributes, such as changes in nighttime lights over time. By the end, readers will be equipped with the skills to create and interpret complex RGB composites, providing deeper insights into geographic patterns and changes on Earth's surface.

      AI Generated

      This summary of the content was generated with the help of AI.

      Show original text

      Download PDF-version

   4. Chapter 3. Survey of Raster Datasets

      • Open Access

      Andréa Puzzi Nicolau, Karen Dyson, David Saah, Nicholas Clinton

      This chapter introduces readers to the diverse types of image collections available in Google Earth Engine, including satellite images, composites, land use and land cover maps, and weather data. It provides step-by-step guidance on accessing, filtering, and visualizing these datasets using JavaScript and the Earth Engine API. The chapter also highlights the importance of understanding image collections for performing spatial analyses and offers practical examples and code snippets to enhance learning. By the end of the chapter, readers will have a solid foundation for working with various types of raster datasets in Earth Engine.

      AI Generated

      This summary of the content was generated with the help of AI.

      Show original text

      Download PDF-version

   5. Chapter 4. The Remote Sensing Vocabulary

      • Open Access

      Karen Dyson, Andréa Puzzi Nicolau, David Saah, Nicholas Clinton

      This chapter provides a comprehensive overview of the key characteristics of remotely sensed images, focusing on spatial, temporal, and spectral resolutions. It guides readers through the process of accessing and understanding important image metadata in Earth Engine. Practical examples and code snippets are provided to illustrate how to work with different datasets from various satellite platforms, making it a valuable resource for professionals and specialists in remote sensing and geospatial analysis.

      AI Generated

      This summary of the content was generated with the help of AI.

      Show original text

      Download PDF-version

3. Interpreting Images

   1. Frontmatter

   2. Chapter 5. Image Manipulation: Bands, Arithmetic, Thresholds, and Masks

      • Open Access

      Karen Dyson, Andréa Puzzi Nicolau, David Saah, Nicholas Clinton

      The chapter delves into the manipulation of satellite images using Earth Engine, focusing on the combination of spectral bands to create indices that highlight specific land covers or uses. It introduces the concept of spectral indices, which are mathematical combinations of bands to distinguish between different types of land cover. The chapter provides a detailed guide on how to calculate and visualize these indices, such as the Normalized Difference Vegetation Index (NDVI) and the Normalized Difference Water Index (NDWI), using arithmetic operations in Earth Engine. Additionally, it explores techniques like thresholding and masking to categorize and isolate specific areas of interest within images. These methods are essential for remote sensing applications, including monitoring vegetation health, detecting water content, and differentiating urban areas. The chapter also includes practical assignments and code examples, making it a valuable resource for both beginners and experienced practitioners in remote sensing.

      AI Generated

      This summary of the content was generated with the help of AI.

      Show original text

      Download PDF-version

   3. Chapter 6. Interpreting an Image: Classification

      • Open Access

      Andréa Puzzi Nicolau, Karen Dyson, David Saah, Nicholas Clinton

      The chapter delves into the fundamental concept of image classification in remote sensing, focusing on the categorization of land cover and land use. It guides readers through the processes of supervised classification, including training data collection, classifier selection, and image classification. Additionally, it explores unsupervised classification techniques, such as k-means clustering, and compares the results of both methods. The chapter also provides practical assignments to enhance the classification outcomes and highlights the importance of accuracy assessment. This comprehensive approach makes it an essential resource for professionals seeking to advance their skills in remote sensing and Earth Engine.

      AI Generated

      This summary of the content was generated with the help of AI.

      Show original text

      Download PDF-version

   4. Chapter 7. Accuracy Assessment: Quantifying Classification Quality

      • Open Access

      Andréa Puzzi Nicolau, Karen Dyson, David Saah, Nicholas Clinton

      This chapter delves into the crucial process of accuracy assessment in image classification using Earth Engine. It begins by explaining the importance of quantifying classification quality and introduces key metrics such as the confusion matrix, overall accuracy, and the kappa coefficient. The chapter then guides readers through practical steps to partition data, train a Random Forest classifier, and calculate accuracy metrics. It also discusses hyperparameter tuning and the impact of spatial autocorrelation on classification accuracy. By the end of the chapter, readers will understand how to evaluate the performance of their classifiers and identify areas for improvement, making it a valuable resource for anyone seeking to enhance the accuracy of their remote sensing projects.

      AI Generated

      This summary of the content was generated with the help of AI.

      Show original text

      Download PDF-version

4. Advanced Image Processing

   1. Frontmatter

   2. Chapter 8. Interpreting an Image: Regression

      • Open Access

      Karen Dyson, Andréa Puzzi Nicolau, David Saah, Nicholas Clinton

      This chapter delves into the application of regression to interpret imagery, specifically focusing on predicting numeric variables for pixels. It begins by differentiating regression from classification, emphasizing the use of regression for numeric dependent variables. The chapter then introduces the concept of reducers in Google Earth Engine, which are tools for summarizing data over space, time, and bands. Practical examples include linear fit and linear regression reducers, as well as nonlinear regression using the CART algorithm. The chapter also covers the assessment of regression performance through RMSE, providing a comprehensive guide to implementing and evaluating regression models in remote sensing. By the end, readers are equipped with the knowledge to apply these techniques to their own projects, making it a valuable resource for those looking to advance their skills in remote sensing analysis.

      AI Generated

      This summary of the content was generated with the help of AI.

      Show original text

      Download PDF-version

   3. Chapter 9. Advanced Pixel-Based Image Transformations

      • Open Access

      Karen Dyson, Andréa Puzzi Nicolau, Nicholas Clinton, David Saah

      The chapter delves into advanced pixel-based image transformations, emphasizing the use of linear algebra and matrix algebra to extract more information from remote sensing images. It covers tasseled cap transformations, principal component analysis, and spectral unmixing, demonstrating how these methods can improve classification, time-series analysis, and change detection. Practical examples and code snippets are provided, enabling readers to apply these techniques using Earth Engine. The chapter also highlights real-world applications, such as detecting forest degradation and mapping cropping patterns with high accuracy.

      AI Generated

      This summary of the content was generated with the help of AI.

      Show original text

      Download PDF-version

   4. Chapter 10. Neighborhood-Based Image Transformation

      • Open Access

      Karen Dyson, Andréa Puzzi Nicolau, David Saah, Nicholas Clinton

      This chapter delves into the application of neighborhood-based image transformations in remote sensing, focusing on techniques that utilize a spatial component to inform pixel values. It covers essential operations such as smoothing, edge detection, and morphological processing, all of which are foundational in remote sensing workflows. The chapter offers practical examples and code snippets using Google Earth Engine, enabling readers to implement these transformations effectively. Additionally, it explores texture analysis and spatial statistics, providing a comprehensive guide to enhancing image analysis techniques in remote sensing.

      AI Generated

      This summary of the content was generated with the help of AI.

      Show original text

      Download PDF-version

   5. Chapter 11. Object-Based Image Analysis

      • Open Access

      Morgan A. Crowley, Jeffrey A. Cardille, Noel Gorelick

      The chapter delves into the method of object-based image analysis, emphasizing the use of the Simple Non-Iterative Clustering (SNIC) algorithm in Google Earth Engine. Traditional pixel-based classification techniques often suffer from unwanted noise, which object-based methods aim to address by segmenting images into meaningful objects. The SNIC algorithm is particularly useful for mapping forest disturbances and other land use changes by integrating additional context into the classification process. The chapter guides readers through the practical application of SNIC, including unsupervised classification and the detection of objects in imagery. It highlights the advantages of object-based analysis in reducing noise and enhancing the clarity of landscape assessments. Additionally, it explores the effects of varying SNIC parameters on classification results, encouraging readers to experiment and optimize their own analyses.

      AI Generated

      This summary of the content was generated with the help of AI.

      Show original text

      Download PDF-version