User:Mav/Data Acquisition in GIS notes by maveric149/2002-02-22 Lecture
Data Acquisition in GIS notes by maveric149 2002-02-22 Lecuture
GIS Design Model
Getting info, organize info, analyze info
Moing from he real world to a specification to a database (working from a conceptual to an operational model)
Raster
First widespread usage was with mapping and displaying images on TV screens.
- It has a value for each pixel on the screen.
- rid/lattice
Vector
- It is faster for logical expressions.
- Mathematical relationships built of "nodes" and "lines"
- The only thing stored is the w:coordinates
Triangulated Irregular Network (TIN)
Series of triangles that represent an area/polygon.
Coverage
Region
Example of having several islands/polygons, but one state/region (exm: Hawaii). The state would be the "region".
- Used alot in real life; things overlap.
Grids
They are different levels of a raster. Each of which tracks a certain types of raster values (such as vegetatin, elevation, Buildings)
Discrete data
There is only one value for a given cell. It is unique (roads, buildings, water bodies etc.)
Continuous data
Called grids or lattices usually floating point format (elevation, pollution, vegetation etc.)
String attributes in string data
Point, linear, continuous types of data
Normalization
Process of developing rules that break down the data into managable format. (process of simplifying data into a database, in order to make data management easier)
- contains repeating groups
- Owner attributes imply order
First normal form from C.J. Date (circa 1964):
- put the data in columns and rows and put only one value in each cell.
- EXM: record numbers for rows, different categories for column and values in the cells
Primary Keys
In the previous example, the key would be the first row with the numbers; it defines what are the unique IDs.
- If data in database changes dramatically enough, it will necessitate the changing of the key.
Good Primary Key is...
- guaranteed unique
- never changes
- Doesn't contain embedded informaiton
- is never reused
- is available when you need it
- generated keys satisfy the above requirements
Generation of unique IDs becomes an administration issue when dealing with multiple user that need to generate them at the same time.