Raster data model and
comparisons with vector
Representing Spatial Elements
Raster
Stores images as rows and columns of numbers with a Digital
Value/Number (DN) for each cell
Units are usually represented as square grid cells that are
uniform in size
Data is classified as
“continuous” (such as in an
image), or “thematic” (where
each cell denotes a feature type.
Numerous data formats (TIFF,
GIF, ERDAS.img etc)
Raster Data Types
Raster data represent features as a matrix of cells within rows
and columns in continuous space
These cells are of a specific dimension size
Each raster data layer represents one attribute (e.g. elevation)
Image-based Data
Image data ranges from satellite images and aerial
photographs, to scanned maps that have been
converted from printed to digital format
Difference between an image and a grid
• Though both are raster data models
• The shape of all raster data models can either be
rectangular or square
Characteristics Image Grid
Unit Pixel Cell
Value Only positive Both positive and
integers negative integers and real
numbers
Spatial resolution
The measure of how closely pixels can be
resolved in an image is called spatial
resolution, and it depends on properties of
the system creating the image.
For practical purposes the clarity of the
image is decided by its spatial resolution.
Vector (discrete)
and
Raster (continuous) Models
GIS works with two fundamentally different types of
geographic data models:
Vector (discrete) and Raster (continuous)
In the vector model, information about points, lines
and polygons is encoded and stored as a collection
of x,y coordinates
The vector model is extremely useful for describing
discrete features, but less useful for describing
continuously varying features such as soil type
The raster model has evolved to model such continuous
features
Both models have unique advantages and disadvantages
Modern GISs are able to handle both models
Vector vs. Raster
Vector Advantages
Accurate positional information that is best for storing discrete thematic features
(e.g., roads, shorelines, sea-bed features.
Compact data storage requirements
Can associate unlimited numbers of attributes with specific features
Raster Advantages
The most common data format
Easy to perform mathematical and overlay operations
Satellite information is easily incorporated
Better represents “continuous”- type data
Sl. Characteristic Vector Structure Raster Structure
No.
1. Data structure Complex Simple
2. Ease of learning Difficult – software is complex Easy - functions tend to be more
intuitive than in vector
3. Positional precision Can be very precise and thus Precision increased with increased
accurate processing time and data storage
needs accuracy. Limited by pixel
size
4. Attribute precision Good for polygon, point and line Good for continuous data; limited by
data; not good for continuous size of pixels in representing
data unless connected to TIN or attribute distribution in real world
similar technology
5. Comprehensiveness Good for spatial query and Not good for spatial query but very
of analysis capability relatively simple data, analysis- good for spatial analysis filtering,
limited to Intersections and modeling
Sl. Characteristic Vector Structure Raster Structure
No.
6. Overlay ability Good, but overlaying many layers Because all pixels line up, overlay
can cause many splinters, etc. in the procedures do not create problems
result which are difficult to eliminate
7. Storage requirements Relatively small but complex Relatively large and simple but may be
complex
8. Ability to work with Poor - data must be vectorized first Good - uses same kind of data structure
image data
9. Conversion to other map Usually included in package and Difficult and quite often creates warped
projections relatively simple to do images which do not fill the raster,
causing problems with neighborhood
functions
10. Ability to work with Good - because system can handle Poor - raster structure not amenable to
network data structures lines network
11. Cost Expensive Inexpensive
12. Output map quality Very good - looks like a map Poor - doesn't look like a map to lay
people
Vector to Raster Conversion
L = f1 (X, Y)
E = f2 (X,Y)
Raster to Vector Conversion
X = f3 (L, E)
Y = f4 (L,E)
Where,
L = scan line
E = element / pixel number within scan line
X = horizontal coordinates of the GIS system map projection
Y = vertical coordinates of the GIS map projection
f1…f4 = Polynomial functions derived by least squares analysis of control
points which can be identified on an image
THANKS
