Raster Data Model

The raster data model represents geographic data in the form of a grid
made up of rows and columns (cells or pixels). Each cell holds a value
representing some attribute or feature (like temperature, elevation, land
cover, etc.).
Point: Represented by a single cell in the grid. For example, a weather
station or a tree would occupy one pixel, regardless of its actual size.
Line: Modeled as a sequence of connected cells. A road or river would
be shown as a chain of adjacent pixels forming a linear path.
Area: Depicted by a group of contiguous cells with the same value. For
instance, a forest or lake would be represented by a block of neighboring
pixels sharing the same attribute

◆ Features:

• Simple structure: Easy to understand and store.

• Efficient processing: Suitable for large area analysis and continuous
data (e.g., satellite imagery).
• Good for high-resolution data: Especially when features vary
smoothly over space (like soil moisture or pollution levels).

Raster data structures and its types.

Raster data represents geographical data using a grid of rectangular cells
(pixels), where each cell holds a value representing information such as
elevation, land cover, or temperature.
Components of Raster Data:
1. Entity Model
o Represents whole data visually.
o Example: Bare land is shown in dark cells, trees in white cells.
o Easy to understand as a sketch or grid representation.

2. Pixel Values
o Data is numerically encoded. Example:
▪ 1 = Bare land
▪ 0 = Trees
o This converts the entity model into binary/matrix form.

3. File Structure
File structure defines how raster data (a grid of cells) is stored and
organized in a digital format. It ensures that spatial data is efficiently
encoded, stored, and retrieved for processing in GIS and remote sensing
applications.
Types of Raster Data
1. Run-Length Encoding (RLE)
• Stores consecutive cells with the same value as a single value and count.
• Works best when the raster has repetitive data (e.g., large areas of the
same land type).

Example:

2. Quad Tree Encoding

• Most used and efficient method.
• Uses recursive decomposition:
o Divides raster into quadrants.
o If all cells in a quadrant have the same value, it's a leaf node.
o If not, it keeps subdividing and creates branch nodes.

Advantages:
• Reduces storage.
• Efficient searching and indexing.

Structure:
3. Chain Encoding
• Defines boundaries of an object.
• Encodes the movement direction from cell to cell along the boundary.

• Directions:
o 0 = North
o 1 = East
o 2 = South
o 3 = West

Structure:
• First line: Starting point (e.g., (4,2)).
• Next: Series of direction and steps (e.g., 2,2 means move South 2 cells).
Use case: Efficient for storing shape outlines.

4. Block Encoding
• Divides raster into blocks instead of individual cells.
• Blocks are labeled with size and starting coordinates.
• Useful when entire blocks share the same value.
Structure:
• Stores:
o Block size
o Number of blocks
 o Starting coordinates (top-left of the block)

Example:

• Block size 2, starts at (2,2)

• Block size 4, starts at (3,2)

Data Conversion and Integration in GIS

In GIS, spatial data is collected from various sources, often in different formats.
To use these datasets together for analysis or mapping, they need to be
converted into a common format or structure and then integrated

“The process of transforming spatial data from one format or model to another
and combining them for common spatial analysis, regardless of how they were
originally acquired.”
Types of Data Conversion:
A. Format Conversion:
• Converting between file formats (e.g., from CAD .dwg to GIS shapefile
.shp, or from GeoTIFF to JPEG).
• Used when different GIS software or data sources use incompatible
formats.
B. Model Conversion:
• Changing data from vector to raster or vice versa.
• Required when combining raster data (like satellite images) with vector
data (like land parcel maps).
C. Coordinate Conversion:
• Transforming data from one projection or coordinate system to another
(e.g., UTM to geographic coordinates).
• Essential for accurate spatial overlay and measurements.
D. Attribute Conversion:
• Modifying or translating attribute data types or values (e.g., coded
domain values to readable formats).
Rasterization:

Converts vector data (points, lines, polygons) into raster format. Each feature is
assigned to grid cells based on spatial location.
Description:
• Converts points, lines, and polygons into a grid of cells (raster).
• Each cell gets a value representing the feature it covers.
Process:
• Define cell resolution (e.g., 10m × 10m).
• Overlay vector features on the raster grid.
• Assign cell values based on feature presence or attribute.

Example: Converting road networks (lines) into a raster showing road density.

Vectorization:

Converts raster data into vector format by tracing boundaries or extracting

features.
Description:
• Converts grid cells into points, lines, or polygons..
Process:
• Identify connected cells of similar value (e.g., all blue = water).
• Trace boundaries to form polygons or lines.

Example:
• Extract lake boundaries from a classified satellite image
What Is Data Integration?

Data integration is the process of bringing together spatial data from

different sources, formats, and models to enable combined spatial analysis.
In real-world GIS applications, integration is essential because:
• Remote sensing data is often raster-based
• Survey, GPS, and administrative data is often vector-based
• Combining both gives more complete and meaningful results

1.Overlay Analysis
Purpose: Combine raster and vector layers to analyze spatial relationships.
How it works:
Overlay analysis superimposes multiple datasets (e.g., land use raster and
administrative boundaries vector) to identify correlations or patterns. For
example, you can overlay a land cover raster with village boundaries to assess
how much forest area falls within each village.
Use cases:
• Suitability analysis (e.g., best location for agriculture)
• Environmental impact assessments
• Urban planning
2. Zonal Statistics
Purpose: Summarize raster values within vector-defined zones.
How it works:
A vector layer (like district polygons) defines zones. The raster layer (e.g.,
NDVI or rainfall) provides values. The tool calculates statistics (mean, sum,
min, max) for each zone.
Use cases:
• Average rainfall per watershed
• NDVI trends per village
• Elevation range within protected areas
3. Sampling
Purpose: Extract raster values at specific vector locations.
How it works:
You use vector points (e.g., weather stations) or lines (e.g., roads) to sample
values from a raster (e.g., temperature, elevation). Each point or line segment
gets the corresponding raster value.
Use cases:
• Elevation at road intersections
• Soil moisture at farm locations
• Pollution levels along a river
4. Rasterization
Purpose: Convert vector data into raster format.
How it works:
Each vector feature (point, line, polygon) is assigned to raster cells based on
spatial location. Attributes can be preserved as cell values.
Use cases:
• Converting road networks into a raster for cost-distance analysis
• Creating a raster mask from administrative boundaries
• Modeling fire spread using rasterized vegetation zones
5. Vectorization
Purpose: Convert raster data into vector format.
How it works:
Raster cells with similar values are grouped and traced into vector features
(points, lines, or polygons). This is useful for editing or querying.
Use cases:
• Extracting land cover boundaries from classified satellite images
• Converting elevation contours from DEMs
• Digitizing features from scanned maps
6. Geo-referencing
Purpose: Align raster and vector data to a common coordinate system.
How It Works: Known ground locations (like road junctions, schools) are
matched with points on the raster (like a scanned map). These control points
help “stretch” and align the image properly.
Real-Life Example:
A city planner scans a 1970 land use map and geo-references it using road
intersections to compare it with today’s vector road map, helping them analyze
urban sprawl over time.
Use cases:
• Aligning historical maps with modern GIS data
• Registering satellite imagery to a base map
• Integrating scanned cadastral maps with parcel boundaries
 Summary Table
Technique Purpose Real-Life Example
Overlay Analyze spatial
Forest area inside each village
Analysis relationships
Zonal Calculate statistics
Avg rainfall in each district
Statistics in zones
Sampling Get raster values at Elevation at road intersections
vector points
Convert vector → Rasterize roads for travel-time
Rasterization
raster analysis
Convert raster → Extract forest boundaries from
Vectorization
vector satellite images
Match scanned 1970 map to
Geo- Align raster with
today’s GIS layers for
referencing vector maps
comparison

1. Remotely Sensed Data

Remotely sensed data refers to data collected from a distance, typically using
satellite images or aerial photographs. This kind of data plays a vital role in GIS
by offering large-scale, regularly updated views of the Earth’s surface.

Characteristics:
1. Satellite Images as a Data Source:
o GIS projects often begin with satellite images to extract land use,
land cover, or natural feature data.
o Examples: vegetation, crop health, soil erosion, geological
structures, water body depth/composition.
2. Periodic Data Collection:
o Frequent satellite passes allow continuous monitoring of
environmental changes (both terrestrial and aquatic).
3. Digital Orthophoto Quads (DOQs):
o These are geometrically corrected aerial photographs.
o Combine image details of a photograph with the scale accuracy of
a map.
 詍
詏
n 2. Field Data
’˙•·,
詐
詑

Field data is gathered directly from ground-based methods and provides high
accuracy for small-scale projects. It is mainly categorized into:

◆ A. Survey Data
1. Instruments Used:
o Measuring tapes, theodolites, transits, levels, and rods.
o Measurements involve:
▪ Distance (in meters/feet)
▪ Direction (azimuth or bearing)
▪ Elevation (vertical difference)

2. Application in GIS:
o Defines parcel boundaries (land units).
o Data represented as a closed traverse (connected sequence of points
and lines).
o Coordinates used to generate points, lines, and polygons through
COGO (Coordinate Geometry).

B. Global Positioning System (GPS) Data

3. GPS-Based Data Creation:
o GPS receivers collect positional data using signals from multiple
satellites.

4. Process:
o At least 3 satellites are needed to determine a location in 3D space
(x, y, z).
o A 4th satellite is used to correct errors such as timing, signal
interference, and atmospheric distortion.
 5. Outputs:
o Accurate points from single GPS readings.
o Lines from a sequence of GPS points (e.g., roads or trails).
o Polygons from multiple lines (e.g., fields or regions).

6. Used For:
o Mapping roads, property lines, infrastructure.
o Real-time field data collection.

‘\3. Methods for Creating New GIS Data

When data is unavailable, GIS professionals use the following four primary
techniques:

1. Digitization

• Manually tracing map features (like roads, rivers, buildings) from images
or paper maps using a GIS software.
• Converts analog data into digital spatial formats.

2. GPS Surveys
• Collecting georeferenced points, lines, or area data using handheld or
mounted GPS devices.
• Highly accurate and suitable for field mapping.

3. Remote Sensing Interpretation

 • Analyzing imagery to interpret land cover, vegetation types, water
quality, etc.
• Often used for environmental studies and agriculture.

4.Data Conversion
• Transforming non-spatial data (e.g., Excel tables or text reports) into
spatial data.
• Example: Converting a table with village names and coordinates into a
GIS point layer.

What is Metadata?

Metadata is commonly described as:

“Data about data.”
According to M. Anji Reddy, metadata in GIS refers to the descriptive
information that explains:
• What the data is,
• How it was collected,
• Who created it,
• When it was created,
• What its spatial properties are, and
• How it can be used.

Definition from Anji Reddy's Perspective:

“Metadata is the background information about spatial data that helps in
assessing its quality, source, reliability, and applicability for a particular GIS
application.”

What Metadata Typically Includes:

Category Details Stored
Identification Name of dataset, description, keywords
Source Who created the data, when and where it was collected
Spatial Coordinate system, projection, datum, spatial extent
Reference (bounding box)
Data Quality Accuracy, completeness, resolution, limitations
Attribute Info Description of each field in the attribute table
Usage How the data can be used, constraints, license terms

Example (for a land use vector layer):

Metadata Field Example Value
Dataset Name Land Use 2020 – Hyderabad Urban
Creator NRSC (National Remote Sensing Centre)
Date of Creation May 2021
Coordinate System WGS 84 / UTM Zone 44N
Accuracy ±5 meters
Data Type Vector Polygon
Description Classifies urban, forest, agriculture, waterbody
Attribute Info Code: Land use class (1=Urban, 2=Forest, ...)
Usage Constraints Government use only, not for commercial sale

Why Metadata is Important in GIS:

1. Ensures Data Reliability: Users can judge whether data is suitable
for their needs.
2. Improves Data Discovery: Metadata allows easy searching and
cataloging of datasets.
3. Supports Data Reuse: Future users understand the source, scale, and
limitations of the data.
4. Protects Data Integrity: Describes any legal/licensing constraints.

Types of Metadata (as per M. Anji Reddy):

Type Description
Descriptive Metadata Title, abstract, keywords about the dataset
Spatial Metadata Projection, coordinates, spatial resolution, extent
Accuracy, lineage (data creation history),
Data Quality Metadata
consistency
Administrative Creator, data owner, date created, access and license
Metadata terms

In GIS, the version of existing data refers to a specific state or edition of a

dataset at a particular point in time.
According to M. Anji Reddy, as GIS projects are dynamic and updated
frequently, keeping track of versions ensures:
• Data consistency
• Proper update history
• Collaboration without overwriting others' work

In Simple Terms:
Just like software has versions (v1.0, v2.0), spatial datasets also have versions
to track changes, edits, and updates over time.

Why Versioning Is Important:

Purpose Benefit
Track Changes Over Helps users know what edits have been made and
Time when
Supports Multi-user Allows multiple people to work on the same data
Editing safely
You can revert to an earlier version if errors are
Rollback Capability
found
Essential for auditing, legal record-keeping, and
🗂 Maintain Data History
long-term planning

Example:
Imagine a city maintains a building permits GIS layer:
• Version 1.0 (2020): Base data with all buildings approved in 2020
• Version 1.1 (2021): New construction areas added
• Version 2.0 (2022): Building attributes updated with ownership and
permits
Each version captures changes for record-keeping and future analysis.

Reference from Anji Reddy’s Book:

“Versioning is essential in large GIS databases where multiple users interact.
Maintaining different versions of spatial data enables safe editing and ensures
that updates do not overwrite original or base data.”

Related Concepts:
Term Description
The original dataset from which new versions are
Base Version
derived
Child Version A temporary or edited copy made for updates
Version Merging edits from different versions into one
Reconciliation consistent dataset
Archiving Storing older versions for legal or historical use
Steps in the Application of Raster Data Model

1. Data Acquisition
• What happens: Collect raster data from remote sensing sources (like
satellites, drones, or scanned maps).
• Example: Getting a satellite image (e.g., Landsat, Sentinel) of an
agricultural region.

2. Geo-referencing
• What happens: Align the raster data with real-world coordinates so it
fits accurately on a map.
• Example: Adjusting a scanned topographic map so it matches GPS
points on the ground.

3. Raster Data Preprocessing

• What happens: Clean and prepare the raster data for analysis.
• Steps include:
o Noise removal
o Radiometric and geometric corrections
o Image enhancement (contrast stretching, filtering)
• Example: Enhancing the NDVI image to better differentiate between
healthy and dry vegetation.

4. Layer Creation / Classification

• What happens: Convert raw pixel values into meaningful classes (land
use, vegetation, soil types).
• Types:
o Supervised classification (user-defined)
o Unsupervised classification (system-defined)
• Example: Creating a land use map by classifying pixels into urban,
forest, water, etc.

5. Raster Analysis / Modeling

• What happens: Perform spatial analysis using mathematical or logical
operations.
• Common types:
o Map algebra (adding/subtracting layers)
o Slope & aspect analysis (from DEM)
o Suitability modeling
• Example: Finding the most suitable area for farming using slope, soil,
and rainfall raster layers.
 6. Zonal / Statistical Analysis
• What happens: Use vector zones (like district boundaries) to summarize
raster values.
• Example: Calculate average rainfall (raster) within each district (vector).

7. Overlay & Integration

• What happens: Combine multiple raster or raster-vector layers for
integrated analysis.
• Example: Overlay flood risk zones (raster) with village boundaries
(vector) to find affected villages.

8. Visualization & Output Generation

• What happens: Display raster maps with legends, scales, and labels.
Export results for use in reports, web maps, or further GIS.
• Example: Generating a land cover change map and exporting it as a PDF
or GeoTIFF.

9. Storage and Metadata Management

• What happens: Save the raster dataset with proper metadata (coordinate
system, data source, accuracy).
• Example: Storing a classified vegetation map along with its date,
resolution, and satellite source.

Summary Table
Step Description
1. Data Acquisition Collect raster imagery from sensors
2. Geo-referencing Align data to real-world coordinates
3. Preprocessing Clean and enhance the raster data
4. Classification Convert pixels into meaningful land categories
5. Raster Analysis Perform spatial operations and modeling
6. Zonal Statistics Summarize raster data using vector zones
7. Overlay & Integration Combine with other layers for deeper analysis
8. Visualization Create maps and visual outputs
9. Storage & Metadata Save with documentation for future use