Transcript Final Presentation

Raster Database
Group 3
Akash Agrawal and Atanu Roy
1
Chapter Organization
• 1.1 Raster Data
• 1.2 Raster Data in GIS
–
–
–
–
1.2.1 Spatio-Temporal Data
1.2.2 Field Operations
1.2.3 Storage
1.2.4 Retrieval Techniques
• 1.3 Concluding Remarks
2
Learning Objectives
• Learning Objectives (LO)
– LO1 : Learn about Raster Data
– LO2 : Learn about GIS Raster Database
•
•
•
•
•
Why use Raster data in GIS?
How Spatio-temporal data is represented?
What are different Field operations?
What are different Storage techniques?
What are different Retrieval Techniques?
• Mapping Sections to learning objectives
– LO1
– LO2
-
1.1
1.2
3
Raster Data
•
•
•
•
A raster image is rows and columns of cells organized in a rectangular grid.
Each cell is called a Pixel.
Each pixel stores a singular color/attribute value.
Resolution of rater image is denoted by #pixels in row X #column of the
grid.
– 800X600 resolution denotes that the raster image contains 600 rows of 800
pixel each.
4
Learning Objectives
• Learning Objectives (LO)
– LO1 : Learn about Raster Data
– LO2 : Learn about GIS Raster Database
•
•
•
•
•
Why use Raster data in GIS?
How Spatio-temporal data is represented?
What are different Field operations?
What are different Storage techniques?
What are different Retrieval Techniques?
• Mapping Sections to learning objectives
– LO1
– LO2
-
1.1
1.2
5
Raster Data in GIS
• The primary purpose is to display the detailed image on a map area or
render its identifiable objects by digitization.
• Raster maps are ideally suited for mathematical modeling and quantitative
analysis.
• Data storage techniques data are easy to program and gives good
performance for data retrieval.
• Commonly used form of raster data in the field of GIS
– aerial photographs of some area.
• Other raster datasets used in GIS
–
–
–
–
a digital elevation model
Map of reflectance of a particular wavelength of light.
Landsat
Electromagnetic spectrum indicators
6
Learning Objectives
• Learning Objectives (LO)
– LO1 : Learn about Raster Data
– LO2 : Learn about GIS Raster Database
•
•
•
•
•
Why use Raster data in GIS?
How Spatio-temporal data is represented?
What are different Field operations?
What are different Storage techniques?
What are different Retrieval Techniques?
• Mapping Sections to learning objectives
– LO1
– LO2
-
1.1
1.2
7
How Spatio-Temporal data is represented?
• The ST data has become crucial
– to understand cause and effect scenarios
– development of dynamic models for the analysis of it.
• The Snapshot Model
– Every layer in the snapshot model shows the state of geographic distribution
at one time stamp.
– Time intervals between any two layers may vary
– There is no explicit implication for changes within the time lag of any two
layers.
8
Learning Objectives
• Learning Objectives (LO)
– LO1 : Learn about Raster Data
– LO2 : Learn about GIS Raster Database
•
•
•
•
•
Why use Raster data in GIS?
How Spatio-temporal data is represented?
What are different Field operations?
What are different Storage techniques?
What are different Retrieval Techniques?
• Mapping Sections to learning objectives
– LO1
– LO2
-
1.1
1.2
9
Field data
• Field data are an essential part of GIS systems.
– give most up-to-date information about current events
– Needed for creating/updating digital maps
– Help in validating the available data sets.
• Field data source
– Satellites
– Geo-registered sensor networks etc.
• Field data set example
– Satellite images, aerial photographs
– Digitized paper maps
– Earth Science data-sets, e.g. rainfall, temperature maps
10
Field operations
• Field data can be manipulated using
– Map algebra
– Image algebra
• Map algebra vs. Image algebra
– Similarity:
• Operand: raster data
– Difference:
• Image algebra deals with image properties such as color information, number of
pixel, pixel size etc. Example trim/crop, zoom in/out etc.
• Map algebra deals with attribute maps such as temperature map, vegetation map
etc. Example thresholding, gradient etc.
11
Map Algebra
• Map algebra
– Operand: raster data
– Operation: classified in four groups
• Local, focal, global and zonal
• Local operation:
– The value of a cell in the new raster is computed only using the value of that
cell in the original raster.
– Example thresholding, point wise addition etc.
Figure: An example thresholding with threshold value of 4
12
Map Algebra (Cont…)
• Focal operation:
– The value of a cell in the new raster is computed using the value of that cell
and its neighboring cells in the original raster.
– Example focal sum, gradient etc.
Figure: An example of focal operation. (a) Rook neighborehood. (b) Bishop neighborehood.
(c) Queen neighborehood. (d) Focal sum using queen neighborehood.
13
Map Algebra (Cont…)
• Global operation:
– The value of a cell in the new raster is computed using the location or values
of all cells in the original raster data.
– Example: global sum, global average etc.
• Zonal operation
– the value of a cell in the new raster is a function of the value of that cell in the
original raster and the values of other cells which appear in the same zone
specified in another raster.
– Example distance from nearest facility.
14
Image Algebra
• Map algebra
– Operand: raster data/ Image
– Operation:
• ignores the absolute location of pixels.
• come from image processing literature.
• used for display or rendering the image for manual analysis of demonstration
purpose.
• Example: trim/crop, zoom in/out, rotate etc.
Figure: An example trim operation.
15
Learning Objectives
• Learning Objectives (LO)
– LO1 : Learn about Raster Data
– LO2 : Learn about GIS Raster Database
•
•
•
•
•
Why use Raster data in GIS?
How Spatio-temporal data is represented?
What are different Field operations?
What are different Storage techniques?
What are different Retrieval Techniques?
• Mapping Sections to learning objectives
– LO1
– LO2
-
1.1
1.2
16
Storage Techniques
• Traditional Approach
–
–
–
–
standard file-based structure of TIF, JPEG, etc.
use custom software to retrieve data-items of interest
Pros: provide good compression and require less storage space.
Cons: difficult to index the data and hence has slower retrieval operation.
• Database Approach
– stores the raster data items attributes such as geo-location, time-stamp,
various properties etc. in database tables.
– Use database query language such as SQL to retrieve data-item of interest.
– Pros:
• allows quicker retrieval of the raster data.
• allows user defined attributes and support for ad-hoc queries.
– Cons: require storage of millions of significantly sized records.
17
Learning Objectives
• Learning Objectives (LO)
– LO1 : Learn about Raster Data
– LO2 : Learn about GIS Raster Database
•
•
•
•
•
Why use Raster data in GIS?
How Spatio-temporal data is represented?
What are different Field operations?
What are different Storage techniques?
What are different Retrieval Techniques?
• Mapping Sections to learning objectives
– LO1
– LO2
-
1.1
1.2
18
Retrieval Techniques
• Raster data sets are very rich in content
• Retrieval approaches
– Meta-data approach (database approach)
– Content based retrieval (image processing technique)
• Meta-data approach
– stores values of descriptive attributes for each raster data item.
– uses simpler SQL data types such as numeric, string, date etc.
– queries to select a set of descriptive attributes such as location, time-stamp,
subject etc.
– Pros:
• Simpler to implement
• gives accurate answers for queries to select a set of descriptive attributes.
– Cons:
• Queries are limited to descriptive attributes.
• does not support “similarity” based queries
19
Retrieval Techniques (Cont…)
• Content based retrieval or content based image retrieval (CBIR)
– content of an image is represented by extracted primitive visual features such
as representing color, shape and texture.
– Similar image queries are answered based on some combination of these
primitive features.
– CBIR is a two step approach
• Step 1: compute a feature vector or attribute relation graph (ARG) for each image in
the database.
• Step 2: given a query image, compute its ARG and compare to the ARGs in the
database for the image most similar to the query image.
– The success of this approach depends on efficiency of feature and similarity
measure, used to compare two ARGs.
20
21