fbpx
loading
X

Image Analysis with ArcGIS

Overview

Description
Learn best practices and workflows to enhance visualization and extract meaningful information from satellite imagery, lidar, and other remotely sensed data. This course covers dynamic raster processing options available in ArcGIS and takes you on an in-depth exploration of image classification. You will use three classification methods to categorize land cover features and learn how to determine which method is appropriate for a given project and dataset.

Who Should Attend
GIS professionals, image analysts, and others who work with imagery for mapping and analysis. Those working in the forestry, hydrology, environmental management, urban planning, defense, intelligence, and mining industries may find the course of particular benefit.

Goals
After completion of this course you will be able to:
• Apply dynamic raster processing functions to enhance raster display, prepare data for analysis, and quickly create multiple products from a single data source.
• Create a time-series mosaic dataset to visually identify and document areas of change.
• Support change detection, risk assessment, and other types of analysis by performing unsupervised, supervised, and object-oriented classification.
• Assess the accuracy of classification results.

Prerequisites
Completion of ArcGIS 2: Essential Workflows or Using ArcGIS for Geospatial Intelligence or equivalent knowledge

Course Outline

Raster function chains and templates
• Raster functions
• Applying raster functions
• Image Analysis window basics
• Applying raster functions on a four-band raster
• Using a LAS dataset in a mosaic dataset
• Raster function chains
• Reusing raster function chains
• Function template checklist
• Applying a function template to a raster

Visually analyzing change over time
• Determining areas of change
• Sources of raster data
• Improving the display of raster data
• Satellite sensor data exploration
• Identifying areas of change
• Time-series analysis of Ikonos data

Introduction to image classification
• Demand for analytical change detection
• Exploring remotely sensed change
• Image classification history
• Types of image classification
• Classification workflows
• Remote-sensing processing workflow
• Classification outputs
• Planning your analysis

Change detection through unsupervised classification
• Unsupervised classification review
• Characteristics of coarse-resolution data
• Landsat bands
• Exploring Landsat data
• Identifying classes
• Identifying the classes on a Landsat raster
• Comparing results against standards

Supervised classification of developed areas
• Supervised classification review
• Components of a proper training sample
• Creating a spectrally pure training sample
• Evaluating your signature file

Accuracy assessment of classified results
• Using accuracy assessments
• Components of an accuracy assessment
• Accuracy assessment example

Impervious surface analysis with object-oriented classification
• Object-oriented classification review
• Image segmentation
• Segmentation configuration
• Object-oriented training samples
• Creating object-oriented training samples

Applications of image analysis
• Analysis application
• Recommended workflows for the activity