18-11-2015

Two fundamental ways of representing

geography are discrete objects and fields.

The discrete object view represents the real world as

Raster Data objects with well defined boundaries in empty space.

(x1,y1)
Points Lines Polygons

The field view represents the real world as a finite number

of variables, each one defined at each possible position.

f ( y)   f ( x , y )dx
x  

Continuous surface

Raster and Vector Data Vector and Raster Representation

Raster data are described by a cell grid, one value per cell of Spatial Fields
Vector Raster

Vector Raster
Point

Line
Zone of cells
Polygon

Numerical representation of a spatial surface (field) Six approximate representations of a field used in GIS

Grid

Regularly spaced sample points Irregularly spaced sample points Rectangular Cells

TIN Contour and flowline

Irregularly shaped polygons Triangulated Irregular Network (TIN) Polylines/Contours

from Longley, P. A., M. F. Goodchild, D. J. Maguire and D. W. Rind, (2001), Geographic Information
Systems and Science, Wiley, 454 p.

1
 18-11-2015

A grid defines geographic space as a matrix of identically-

sized square cells. Each cell holds a numeric value that
measures a geographic attribute (like elevation) for that unit The grid data structure
of space.

• Grid size is defined by extent, spacing and no data

value information
• Number of rows, number of column
• Cell sizes (X and Y)
• Top, left , bottom and right coordinates
• Grid values
• Real (floating decimal point)
• Integer (may have associated attribute table)

Definition of a Grid Points as Cells

Cell size

Number
of
rows
NODATA cell
(X,Y)
Number of Columns

Line as a Sequence of Cells Polygon as a Zone of Cells

2
 18-11-2015

NODATA Cells Cell Networks

Grid Zones Source of raster data

• Data
• digital elevation
• satellite images
• digital orthophotos
• scanned maps
• graphic files

Elements of Raster Data Model Elements of Raster Data Model

• Grid, raster map, surface cover, or image • Can have value = “pointer”
• Rows and columns with cells • Cell size determines resolution
• Origin at upper left, 2 dimensional coordinate • Size vs resolution
system • Projected on coordinate system
• Value = integer or floating point • Can be displayed with vector if in same coordinate
• Usually codes for interpretation system
• No use for DBMS because grid is both spatial and • Easier to do data manipulation, aggregation and
attribute data analysis

3
 18-11-2015

Types of Raster Data Types of Raster Data

• Satellite Imagery • Satellite Imagery
• remotely sensed • Commercial
• Landsat multispectral, 79 meters, ’72-’92 • Space Imaging 1 meter
• Landsat 4, Thematic Mapper, 30 meters, • Earth Watch Incorporated
• National Oceanic and Atmospheric Administration • Image Processing Packages
(NOAA) Advanced Very High Resolution Radiometer
(AVHRR) • ERDAS
• French SPOT 1986 • ER Mapper
• India, Japan • ArcView

Digital Elevation Models Digital Elevation Models

• DEM is point based with elevation at center of a • Non-USGS DEMS
cell. • SPOT
• Each file contains • Intermap Technologies (5-10 m x 2-5 m)
• Elevation, • Global DEMs
• Header: units, min/max elev, proj, accuracy • GTOPO30
• Four types • ETPOPO5
• 7.5 minute DEM (30 or 10 meter elev, 4 levels)
• 30 minute DEM (60 meter)
• 1 degree DEM (100 meter)
• Alaska DEMs

Digital Orthophotos Other

• Digital Orthophoto Quad • Binary scanned files
• Georeferenced • Digital Raster Graphics (DRG) 8 ft
• 3.7 minute DOQQ 1m x 1m • Graphic
• Combines image characteristics of a photograph • TIFF
with the geometric qualities of a map. • GIF
• JPEG
• Ideal background
• GeoTIFF
• Proprietary: grids or raster maps

4
 18-11-2015

Scanning
of images Raster Data structure
• How to store raster values in digital format,
so that it can be used and processed by
computers.
• Cell by cell encoding
• Run length encoding
Figure. Small format scanner.
• Quad tree

Raster Data structure … Raster Data structure …

Runlength Compression (for single layer)

Cell by Cell encoding Full Matrix--162 bytes Run Length (row)--44 bytes
• Simple raster data structure 111111122222222223 1,7,2,17,3,18
• Raster stored as matrix (row & Columns) 111111122222222233 1,7,2,16,3,18
• Ideal for continuous change in cell value viz., 111111122222222333 1,7,2,15,3,18
DEM 111111222222223333 1,6,2,14,3,18
• DEM use this because neighboring elevation 111113333333333333 This is a “lossless”
1,5,3,18
values are rarely the same. 111113333333333333 compression, as 1,5,3,18
opposed to “lossy,”
111113333333333333 1,5,3,18
• Satellite data (one layer & multilayer since the original

concepts) 111333333333333333 data can be exactly 1,3,3,18

reproduced.
• BSQ 111333333333333333 1,3,3,18
• BIL • Cell by Cell encoding: insufficient of a raster contains many redundant Cell value
• BIP • RLE: effectively store repetitive values, by row and by group
• RLE: encoding & compression method
• GRASS, IDRISI, ArcGIS

Quad Tree Further readings

• Uses recursive
• Kang-tsung Chang (2007), ‘Introduction to Geographic
decomposition to divide a Information Systems’ Tata McGraw Hill, New Delhi.
raster into hierarchy of
• [Link] and Albert [Link] (2006) “Concepts and
quadrants. Techniques of Geographic Information Systems”
• Recursive decomposition Prentice Hall of India,New Delhi.
refers to a process of • Burrough, Peter A. and Rachael McDonnell, (1998),
continuous subdivision ‘Principles of Geographical Information Systems’ Oxford
until every quadrant in a University Press, New York.
quad tree contains only • Magwire, D. J., Goodchild, M.F. and Rhind, D. M.,
one cell value. (2005), ‘Geographical Information Systems: Principles
and Applications', Longman Group, U.K.