Advanced Image Management using the Mosaic Dataset

1
(No Transcript)
2
Advanced Image Management using the Mosaic
Dataset

• July 25, 2012

• Vinay Viswambharan, Mike Muller

3
Agenda

• ArcGIS Image Management

• Concepts and Terminology
• Image Management Challenges
• Image Management using ArcGIS
• LiDAR Data Management using ArcGIS
• Esri Projects

4
Concepts and Terminology
5
What is Imagery in ArcGIS?

• ArcGIS is the leading Geographic Information
  System
• ArcGIS has supported Raster Data for decades
• Visualization, Analysis, Data Management
• Imagery is Raster Data
• Imagery is part of the core information model in
  ArcGIS

• Imagery / Raster Data are not new to ArcGIS

6
Imagery is Raster Data

• ArcGIS supports imagery
• Aerial Photography
• Satellite Imagery
• Thermal Imagery
• Digital Elevation Models
• Scanned Maps
• RADAR
• Thematic Images (e.g. classification results with
  attributes)
• Analysis Surfaces (e.g. interpolated surfaces
  from points)

• ArcGIS works with all types of Raster Data

7
History of ArcGIS and Imagery

• ArcGIS continues to evolve as an imagery system

8
Describing the Capabilities Trend

• When we think of imagery, we think sensor
• A more capable system needs a rich information
  model to be more intelligent with sensor data
• Advancements in the ArcGIS imagery information
  model describe this capabilities trend

Information Model Concept ArcGIS Version
Raster Dataset 5 (Arc/Info), 8.0
Raster Type 10.0
Raster Function 10.0
Raster Product 10.1
Mosaic Dataset 10.0
Image Service 9.2 (classic), 9.3 (ArcGIS for Server)

• ArcGIS 10.1 has a comprehensive imagery
  information model

9
Raster Dataset

• Primary information model to read and write
  raster storage (pixels)
• Comprehensive
• 1 or N Bands
• 1-64 bits per band
• compressed or uncompressed
• 80 formats supported
• pyramids (rrd, ovr, internal)
• Read directly no need to convert

• ArcGIS supports working with individual images

10
Raster Type

• Defines rules for ingesting imagery
• Sensor specific
• Defines processing chains
• Defines pixel storage and metadata schema
• Defines georeferencing

11
Raster Function

• A function that can be applied to a pixel or
  block of pixels
• Machinery behind on-the-fly processing
• Chained together to create processing chains
• Geometric or Radiometric
• 38 out-of-the box
• Developers can extend ArcGIS image processing by
  implementing a Raster Function

• ArcGIS processes imagery on-the-fly

12
Raster Product

• Easy to use shortcuts to well known band
  combinations and processing chains
• Users dont need to understand all the
  constituent files associated with an image
  product

• ArcGIS makes it easier to work with well known
  image products

13
Mosaic Dataset

• A mosaic dataset is a geodatabase data model for
  the management and access of small to vast
  collections of imagery (raster data)
• A catalog of imagery that stores references to
  imagery, while ingesting and storing all
  associated metadata
• Flexible dynamic mosaicking
• on-the-fly processing chains

• ArcGIS image management is centered around mosaic
  datasets

14
Image Service

• Provides access to imagery (raster data) through
  a Web Service
• Web enabling imagery
• single images
• large collections using mosaic datasets
• Support dynamic processing
• Standards - Esri, WMS, WCS, KML
• APIs - REST, SOAP, ArcObjects

• Powered by ArcGIS for Server

15
Image Service

• More than a picture or basemap

• Access imagery as an image (visual analysis)
• Access imagery as raster data (pixel analysis,
  DNs)
• Access imagery as a catalog (collection with
  metadata)

16
Image Management Challenges
17
Image Management Challenges

• Multiple Formats
• Multiple Sensors
• Processing requirements vary
• Products delivered as collections of files
• Metadata organization and format
• Storage
• Massive collections
• Dissemination of derived products

18
ArcGIS Addressing Image Management Challenges
Raster Dataset

• Multiple Formats
• Multiple Sensors
• Processing requirements vary
• Products delivered as collections of files
• Metadata
• Storage
• Massive collections
• Dissemination of products

Raster Type
Raster Function
Raster Product
Raster Type, Mosaic Dataset
Mosaic Dataset
Mosaic Dataset
Image Service
19
Image Management using ArcGIS
20
Image Management Tools

• Built on ArcGIS Geoprocessing
• More than 80 tools for image management and
  processing
• Authoring Mosaic Datasets and Image Services
• Format Conversion
• Compression
• Statistics Calculation
• Pyramid Building
• Clipping / Spatial / Spectral Subsetting
• More

• ArcGIS Geoprocessing is a proven processing
  framework for geospatial data

21
Mosaic Dataset

• Dynamic Mosaicking

• By default users see best available imagery
• Fuse of overlapping imagery from multiple sources
• User control of image order
• User control of resolving overlapping pixels
• on-the-fly mosaicking

22
Mosaic Dataset

• Dynamic Mosaicking

• Mosaic Methods
• Defines the order in which images are selected
• Closest to Center closest to the view center are
  placed on top
• Closest to Nadir distance between the nadir
  position and view center
• Closest to Viewpoint North, South, East, West
• By Attribute ORDER BY Sun Angle, ORDER BY
  Acquisition Date
• NorthWest centers to the northwest are displayed
  on top
• Seamline feathered blend using a predefined
  seamline shape
• Lock Raster display a single or multiple rasters
  using identifier
• None order in the catalog

23
Mosaic Dataset

• Dynamic Mosaicking

• Mosaic Operators
• Defines how overlapping pixel are resolved

FIRST
MIN
LAST
MAX
BLEND
MEAN
distance weighted algorithm
24
Mosaic Dataset

• on-the-fly processing

• Raster Types provide default processing
• Author can build processing chains
• Raster Functions chained together
• Can be applied to
• Mosaic Dataset
• Raster(s) in a Mosaic Dataset
• Raster(s) in Desktop

25
Mosaic Dataset

• on-the-fly processing

• Non-destructive
• No need to alter source images or create derived
  intermediate images
• Create multiple products from the same source
  images without storage costs of intermediate
  imagery

26
Mosaic Dataset

• Solves image management challenges

• Time
• Ingest imagery closer to the sensor / time of
  acquisition and process on-the-fly
• No need to alter source images or create derived
  intermediate images
• Money
• Reduced storage costs
• Easier to manage large collections (automate
  maintenance)
• Value
• Create multiple derived products from a single
  source
• Catalog your withholdings
• Easy access to metadata
• Manage large multi-resolution, multi-sensor,
  disparate datasets in a single collection
• Processing results are immediately visible
  without having to wait for persistent
  representations

27
Mosaic Dataset Demonstration

• Vinay Viswambharan

28
LiDAR Data Management using ArcGIS
29
Managing Large Collections of LAS

• LAS files can be ingested into a Mosaic Dataset
• using LAS Raster Type
• configurable (e.g. cell size, return to use, void
  filling)
• Users work with the raster representation or
  download the source LAS files

30
LAS Data Management Workflow
Visualization Analysis
Large Collections of LAS files
Raster Data Consumers
QC
Mosaic Dataset
Visualization Analysis
Share as Image Service
Data Managers
Download Project LAS files
LAS Dataset
Directly accessed Project LAS files

• Visualize
• Analyze
• Edit / Reclassify / Constraints
• Geodatabase Terrains

LAS Data Consumers
31
Mosaic Dataset LAS Demonstration

• Vinay Viswambharan

32
Esri Projects
33
Landsat Services

• Access to USGS Global Land Survey for ALL EPOCHS
• 1975, 1990, 2000, 2005, 2010 new
• 50 Dynamic Image Services
• not just tiled basemaps
• 10 TB of imagery
• FREE!
• http//www.esri.com/landsat
• Powered by ArcGIS 10.1

• Esri is hosting imagery using ArcGIS 10 10.1

34
IKONOS World Imagery Updates

• Project Scope
• 50 million sq. km. of imagery
• Select Best Of 2 million IKONOS scene archive
• Select -gt Ortho -gt Color Balance -gt Mosaic -gt
  Cache to Basemap
• 300k scenes processed and growing
• Total Raw Data 152 TB

• Esri is hosting imagery using ArcGIS 10 10.1

35
Coming Soon

• Digital Globe / GeoEye / Esri Partnership
• Best of Hi-Res Satellite
• Aerial 30cm US, 60cm Europe
• ArcGIS Online imagery content will continue to
  grow

• Esri is hosting imagery using ArcGIS 10 10.1

36
Steps to evaluate UC sessions

• My UC Homepage gt Evaluate Sessions
• Choose session from planner
• OR
• Search for session

• www.esri.com/ucsessionsurveys

37

• Thank you for attending
• Have fun at UC2012
• Open for Questions
• Please fill out the evaluation
• www.esri.com/ucsessionsurveys
• Offering ID 881