Raster data model

Also called: Raster format, Grid format · Topic: Remote Sensing, GIS and GNSS · NCERT: Class 12, Ch 4 "Spatial Information Technology"

Meaning

The raster data model is a way of storing map data in a GIS (Geographic Information System, a computer system that stores and studies data linked to places) as a grid of equal cells, also called pixels. Each cell has a row-column position and holds one value, such as brightness, land-use class or height. It matters because satellite images, aerial photographs and scanned maps are always raster. So every remote sensing study, from flood mapping to a DEM, starts with raster data.

Explanation

How it works: cells, positions and values

• A GIS stores the location part of data (spatial data) in one of two forms, called data models: raster or vector.
• In raster, the area is cut into a grid of equal squares (cells).
• The row-column position tells you where a cell sits in the grid.

• Each cell holds one value, for example:

  • Brightness (in a satellite image)
  • Land-use class (forest, crop, water)
  • Height (in a DEM, a Digital Elevation Model: a raster where each cell stores ground height)

• Descriptive information (names, population, soil type) is kept separately as attribute data in tables.

Resolution: cell size decides detail

• The cell size sets the resolution.
• Small cells give fine detail, but there are more cells to store and files get larger.

• Large cells give rough detail, but files are smaller.

• Curved or slanted edges, like a coastline, look like "stair-steps" because they are built from square cells.

Advantages and disadvantages (NCERT Box 4.2)

• Advantages:
• Simple data structure: just a grid of numbers.
• Easy overlay: two grids of the same cell size can be laid over each other and compared cell by cell.
• Fits satellite images: images already come as pixels.

• Handles high spatial variability (values that change a lot from place to place).

• Disadvantages:

• Uses a lot of storage.
• Shapes look rough (stair-steps), so map output is less precise.

• Poor for network analysis (roads, pipelines) and projection change (redrawing the map on a different map projection).

• Best for: continuous variables, meaning things that change smoothly across space: elevation, temperature, rainfall and land use [5].

How raster data is made and converted

• Scanning turns a paper map into a raster image:
• A light source shines on the map.
• Reflected light falls on charge-coupled devices (CCDs), tiny light sensors.

• Each CCD turns light (photons) into electron counts, which are stored as digital values. Together these values make the raster.

• Vector → raster (rasterisation): a grid of chosen cell size is laid over the vector data, and each cell takes the value of the feature under it.

• This is the more common conversion, because most spatial analysis is raster-based.

• Raster → vector (vectorisation): cell edges are traced into lines and polygons, mainly to save storage. The output needs smoothing to remove the stair-steps [6].

Where it occurs

• ISRO's Bhuvan geoportal (run by NRSC, Hyderabad): satellite images, including IRS imagery and terrain data, are raster layers [3].

• On Bhuvan, raster data is served through UMN MapServer and TileCache [4].

• Low-cost backdrop maps: raster images are used as the background layer that roads, boundaries and other vector data are drawn on.

• SVAMITVA: the Survey of India has mapped over 2.8 lakh villages with drones in Andhra Pradesh, Haryana and Karnataka (reported February 2025). The drone images are raster. The property boundaries traced from them are vector polygons [2].
• National Geospatial Policy (NGP) 2022, notified on 28 December 2022, sets targets for a high-accuracy DEM, which is a raster product, by 2030: 25 cm in plains and 1–3 m in hilly and mountainous areas [2].
• World example (USGS, 2002): geologic maps were scanned at 300 dpi, 24-bit colour, cleaned, and then auto-vectorised. The traced lines had a stair-step look with too many vertices and had to be smoothed [6].

Don't confuse with

• Vector data model: stores features as points, lines and polygons using real (x, y) coordinates. It is compact and precise, but it does not fit satellite images. Raster uses a grid of cells and fits images.
• Topology: the recorded connectivity, adjacency and containment between features. It belongs to the vector model, not raster.
• Digitisation: tracing map features into x, y coordinates. It is the best way to input vector data. Scanning produces raster data.
• Attribute data: the descriptive tables (names, population). The raster grid is spatial data, even though each cell holds a value.

Prelims Hooks

• Satellite images, aerial photographs and scanned maps are stored as raster. Trap: "Vector is compatible with remote sensing images" is false.
• Raster resolution depends on cell size: smaller cells give finer resolution and larger files.
• Raster is better for overlay and high spatial variability. Vector is better for network analysis and projection change.
• A scanner's CCDs turn photons into electron counts, which are stored as digital values. Scanning gives raster output.
• Vector → raster is the more common conversion, since most analysis is raster-based. Raster → vector is done to save storage.
• On Bhuvan, raster data is served through UMN MapServer and TileCache, and vector data through GeoServer and GeoWebCache [4].

Mains Points

• The right data model shapes the analysis (GS-I):
• Raster suits continuous surfaces: flood extent from satellite images, DEMs and rainfall maps.
• Vector suits networks and boundaries: roads, canals and land plots.

• Disaster mapping and Bhuvan's thematic services use both together [3].

• High-resolution raster data for planning (GS-III):

• NGP 2022 targets a high-accuracy DEM (25 cm in plains, 1–3 m in hills) by 2030 [2].

• This gives a better base for flood modelling, watershed planning and infrastructure design.

• From raster images to land records (GS-III):

• Under SVAMITVA, raster drone images of 2.8 lakh+ villages are turned into vector property boundaries [2].
• Precise records help with land titles, bank loans and fewer property disputes [2].

Related concepts

• Vector data model
• Topology
• Digitisation
• Scanner
• Charge-coupled device
• Georeferencing
• Raster-vector conversion
• Exact matching
• Hierarchical matching
• Fuzzy matching

Read more

• Raster and vector data, data input and matching