User:Mav/Introduction to GIS notes by maveric149/2002-03-19 Lecture
Introduction to GIS notes by maveric149 2002-03-19 Lecture
Raster data structures
[edit]Cector data structre is good for siaplay discrete features, not. Raster data structure is good fo coverages, generates calure for every cell in a grid.
Examples
- elevation data, satellite imagery, orthophotes, scanned maps, graphic files
Way use raster?
- large additonal set of GIS analyses, most GIS software allow simultaneous display a and alysis of reawster and vecotor data.
Componets
[edit]- Grid consistis of rows and columns
- Point is difine by a whole cell in raster model
- line is defined by a sequence of neighboring cells
- Areas are represented by collections of contiguous cells.
Cells
- integer vlues represent categorical data
- floating point numbers typically represtend continuous data
- No spearation between spatial and attrbute data
- resolution is determined by cell size. large cells can't represent the precise locacton of spatial features.
- greid is usually projected on a coordinate system like UTM
- Grid data have fied cell location
Types of raster data
[edit]remote sensed images (Landsate, AVHRR)
- SPOT imagery from France
- all need separtate image processig software (RDAS and ER Mapper)
- Digital Elevation Models (array of uniformly spaced elevation data that is point based)
Digital orthophot quad
[edit]- processed aerial photographs to remove distortion caused by camera tilt and relief
- georeferenced
- can be registerd with topographic can other maps
Digital raster graphics
[edit]- scanned USGS topo map
Graphic fies
[edit]TIFF, GIF, and JPEG
Encoding
[edit]- cell by cell encoding (every cell has a value in the grid) = no compression, used for DEMs
- Run length encoding (only mention cells that have a specific value & then only note the beginning and end values of each row) = some compression of data
- Chain code compression (records the boundary of a region of redundant values by specifyin cardinal direction) = greater amount of compression
- Block code compression (breaks raster region into square blocks) = stars by stating
Compression
[edit]- Lossless or reversible compression
- Lossy or Irreversible compression
MrSID (multi-resolution seamless image datbase) allows ecalliung the image at differeent scales and resutions
- Zip, tar
Distortion
[edit]Warping or rubber sheeting ay be needed to allow the closest fit between raster and vector data (georeferencing)
Conversion
[edit]rasteriation or vectorization of data (both need computer algorihms)
Integration
[edit]allow for processin of sicrete and continuopus data at the same time.