GIS by huangyuarong


									  Geographic Information Systems
                                 M. Neilsen

       GRASS GIS         Quantum GIS      World Wind GIS, NASA       ESRI ArcGIS

* some content is from Grass GIS User’s Guide, Quantum GIS User’s Guide, and Wireless and
GIS, Radim Blažek,, Trento, Italy, and GRASS GIS for Anthropologists by Isaac Ullah

1 Geographic Information Systems (GIS)
    GIS capabilities – GRASS features
    Data structures in GIS
    GRASS vector architecture

2 Vector network analysis
    Shortest path
    Steiner trees
    Allocation of sources

3 From WebGIS to portable GIS
    Portable GRASS
    Linking GRASS to other Free Software

4 Conclusions – GIS and wireless
        Features of a GIS

managing and generating spatial data, usually in a projected coordinate systems

spatial data analysis:
      - map algebra, buffering,
  filtering etc.
      - image processing
      - vector network analysis
      - (geo) statistics
      - much more...

data visualization

                                               Snapshot of a GRASS session
GRASS: Free Software GIS, Quantum GIS – Shareware GUI, ESRI ArcGIS - Commercial

 GRASS is a fully featured Open Source GIS released under GNU General Public License (GPL)
             free availability of source code (1.1 million lines of C code)
             portability: Linux, SUN, MacOSX, MS-Windows, iPAQ, Zaurus, SGI, ...
 stable version 5.0, based on more than 15 years of development
 modular command structure: no monolithic software
 vital development team of around 10 core developers with worldwide contributions


(Germany): Centralized code

                                     Main site at ITC-irst:
                                                 GRASS software web sites:
                                                                              mirrors of
        GIS data structures – Raster

2D raster map: pixels

Data structure:
     x,y,value                                                        Image

                                       Raster map

3D raster volume: voxels

Data structure:

                                        Isosurfaces representation:
                                        relative humidity
        GRASS raster applications

2D raster application

       Interpolation of raster surfaces
       from distributed point data; e.g.,
       soil samples

Available interpolation methods
 bilinear interpolation
 cubic convolution
 inverse distance weighted average (IDW)
 nearest neighbour resampling
 regularized splines with tension (RST)

                                            Field interpolation from local measurements with RST
       GRASS raster applications

2D raster application
      Impact analysis by distribution models

            Noise dispersion
            from highway
        GRASS raster applications

2D raster application

Spatial diffusion algorithms

                               Average territory accession time from main road network
        GIS data structures – Vector

2D vector map

Data structure


                                                                               2D or 3D TIN
3D vector map                                                                  (triangulated
Data structure


                                         3D CAD drawings in GRASS (3D faces)
          GRASS vector data structures

GRASS 5.1 vector geometry types
  line                                                                        node

  boundary                                          vertex

  centroid                                                                           Line
  3D face
  3D kernel                                  node

Vector Architecture Features                                                         Area
  Vector topology is generated                           centroid         vertex
  Spatial index is generated (increased
  access speed)                                                 vertex
  Vector Network library is implemented
                         Vector Data

•   Discreet geometrical objects which are either points, lines, or
•   Vertices are placed by X and Y location for all vector types,
    and shapes are made by geometry
•   For line and polygons, the vertices are joined by lines
    according to a function
•   Attributes are associated with each shape
•   Attributes are stored in a database of info (and therefore each
    object can be multidimensional)
•   Easy database editing with Excel/Open Office (most are in .dbf
                         Raster Data

•   Continuous data (a matrix of values)
•   Each layer has 3.5 dimensions of data
•   Multiple layers can be stacked to represent many dimensions
    of data
•   Display of data can be adjusted and tweaked for heuristic
•   Raster surfaces can be interpolated from discreet (ie. vector)
•   Can map fuzzy datasets, and so can be used to model all sorts
    of non-categorical data
•   Complex statistics and math can be done at each pixel on
    single layers or as functions of two or more layers
                      Which is Better?

•   It depends on your needs
•   Vectors are better for associating many data types with one
    spatial object (i.e., site point) in one file
•   Vectors can only be used to represent discreet phenomena
•   Raster's are better at representing massive amounts of
    spatially differing data
•   They are also better for doing mathematical operations on that
•   They can represent discreet data, but only in one dimension
    per layer
•   Resolution counts! As do extents!
          A Quick Note on Projections

•   Maps are flat representations of a round world

•   Different projections are different ways to
    mathematically “unbend” curvilinear distances into flat

•   Projections also have different Datum point from which
    all measurements are tied back to the Earth

•   While you have absolutely no need to know how or why
    projections work, you should know about two of the
    major types and what the difference between them are.
             A Quick Note on Projections

•   One is the Latitude/Longitude (Lat/Lon) projection, and it works

•   However, all distances in this type of projection are measured
    as fractions of the Earths diameter (degrees, minutes, and
    seconds or decimal degrees)

•   The other is the Universal Transverse Mercator (UTM)
    projection, which is broken up into a series of zones across the

•   It’s units are meters, but you must stay within only the correct
    zone, or your data will become distorted
Lat/Lon Projection
UTM Projection
UTM Zones
         GRASS 5.1 vector architecture

             Geometry                                         Attributes

            Vector library                               DBMI library
Native    SHAPE file   OGR PostGRASS
                                            PostgreSQL    ODBC          mySQL.     DBF

               Files: GML,
  File            Arc,...              PostgreSQL         e.g. Oracle      MySQL         File
      GRASS 5.1 Vector Architecture

Sample Implementation in PostGRASS

                                                                     Table 1 roads
  Table                        Table
                                                                   ID     Attributes
  transport_geometry           transport_category                    7
 Coordinates      FID         FID Field      Cat                    ...
355134 577245      3           3      1       7
355063 577538      5           5      2      13

                                                                   Table 2 crossroads
                                                                   ID         Attributes

                                                    Vector geometry can be linked to
                                                    one or several attribute tables
            Vector map
                                                    via “Field” definition.
          Vector network analysis with Directed Graph Library
The Directed Graph Library: library for vector network analysis

Directed Graph Library provides underlying functionality for GRASS commands
using edge and node weighted, oriented graphs:     shortest path, Steiner trees, allocation of sources


                              B                             Application: shortest path
                                                            find shortest path based on traveling costs:

                                                                  calculated from lines lengths
                                                                  or attributes read from DBMS table(s)


                                                             Shortest path from A to B
       Vector network analysis with Directed Graph Library

Network analysis based on Directed Graph Library

                                                   Application: Minimum Steiner tree

                                                   Subgraph connecting various network nodes
                                                   at minimum costs

                                                    Hospitals connected by broadband cable
       Vector network analysis with Directed Graph Library

Network analysis based on Directed Graph Library

  Allocation of subnetworks to centers

  Travel cost calculation based
   on line length or vector attributes

                                                   Patches of fire brigades in a city
         GRASS and Geostatistics

 GRASS linked to
   R stats and XGOBI

 Data may be kept in database

environment snapshot
       GRASS on the road: Mobile GIS

Port of GRASS to handheld devices: “babyGRASS”

                                                 GRASS on Zaurus

         MapServer: WebGIS

                                MapServer on
                                Mobile Phone

MapServer for data gathering
and dissemination at ITC-irst
        GRASS linked to other Free Software

Interoperability and data accessibility

   Online data dissemination                 (Geo)statistics


             GRASS and Wireless Applications

    What can GRASS provide for wireless applications?

An environment for real data geocoding

Tools for data collection, tools for planning

Storage and management of spatial data with attributes

Data analysis – statistics: interactive and automated

Implementation of predictive models

Data visualization and distribution
          World Wind GIS, NASA

•   Open source, Java GIS Engine and Libraries
World Wind GIS
World Wind GIS

To top