Pitney Bowes
MapInfo Pro™ Monthly Journal
Get on the Grid: An Overview of Raster GIS Concepts and Capabilities.

Easy Tip Thanks to the efforts of our great Engineering team, Pitney Bowes has made available a next generation raster grid based add-on for MapInfo Pro. This includes a new engine and an innovative (and very powerful) grid data format. For those of you who may be familiar with some of our other products in this space, this new capability will be a successor for Vertical Mapper and our Engage 3D products. The new engine and grid data format will also be incorporated into our Discover 3D product line and Spectrum Spatial.

For those of you who may not be too familiar with raster based analysis, I will use this article to introduce you to some of the basic concepts.

Uses for raster grid based analysis

Raster grid based analysis has many uses. We’ll go into a bit more detail next month but for now, here are a few example applications:
  • Wireless telecommunication companies analyse signal strengths along with other data (such as information from drive testing) to plan, design and optimise their wireless network.
  • Public sector organisations at all levels of government make wide use of various environmental data, especially elevation, in a variety of applications. This includes both planning for and responding to emergencies. Analysis of the impact that building and construction may have on the environment and people can also benefit from grid based analysis.
  • Public safety organisations can produce “hot spot maps”, a form of grid map that represents crime or other types of incidents. This can be used for strategic or tactical initiatives to prevent and reduce crime or mitigate other types of incidents (auto accidents, fires, etc).
  • Insurance companies employ grid based analysis, for example, an understanding of elevation to better understand the risk to their portfolio. A low elevation can represent a flood risk while a high elevation might represent a risk from wind damage.
  • Mining and exploration companies employ a variety of grid based analytical techniques in geochemical analysis to add in their efforts to find and exploit natural resources. Check out http://www.pitneybowes.com/pbencom/products/location-intelligence-and-gis/encom-discover.html for more info.
Vector and Raster GIS data

There are two principle forms which the data used in a Geographic Information System (GIS) can take, vector based and raster based. MapInfo Pro (on its own) is primarily a vector GIS. The astute among you will note that within the thematic mapping capabilities, MapInfo Pro can create grid maps. However, MapInfo Pro, on its own, is limited to basic visualisation. Our add-on products such as Vertical Mapper, Discover, Engage 3D, Crime Profiler and the new MapInfo Pro Advanced add raster GIS capabilities to MapInfo Pro that go beyond simple visualisation.

Vector GIS data

Vector GIS data is represented as discrete objects, typically consisting of points, lines and regions. (There are more types but the detail does not need to concern us here.) These objects are associated to attribute data. The attribute data may be contained in a MapInfo table, spreadsheet, database, text file or other file type.

vector data
Above is an example of vector point data overlaid on boundary data . The points in the map represent capital cities. The table on the right is the attribute data associated with the points.

Vector GIS data is not optimal for displaying data that continuously changes through space. Examples of such data include elevation, radio signals, air temperature, noise levels, etc. This is where raster grid based data is advantageous.

Raster GIS data

The word "raster" refers to the image or the visual display component of a map. A raster image can take many forms, such as .bmp, .tif, .jpg etc… If you examine a raster image closely you will notice that they are made up of many (usually) square cells. In a raster GIS, these cells are used to represent geographic data. Every cell contained in the raster map also has either numeric or character attribute information associated with it. A raster based GIS can portray continuously varying data more effectively and can analyse multiple layers of data easily. Raster GIS is relatively fast and you are able to do types of visualization and analysis that are not possible in a vector based system.

raster example
Above is an elevation map. The map on the right is zoomed in very close to show the individual cells. Each cell has an elevation value.

The ideal GIS system combines the best of both worlds. MapInfo Pro, along with our raster/grid add-ons is a very powerful combination.

Understanding grids

The word grid (or “raster grid”) is used to distinguish the difference between a “plain” raster image and an image where each cell has one or more data fields associated with it (aside from the color). In other words, a grid is more than just a picture, it contains information. This information can be queried and analyzed.

Numeric grids

As the name implies, numeric grids use numeric values. This type of grid can be used to represent data that varies continuously across an area. Examples include elevation, air temperature, noise levels, pollution readings, rainfall, the slope (steepness) of terrain, radio signals and geophysical data.

Numeric grids may also be created using other forms of data such as demographic statistics, crime statistics, insurance risks, sales figures or other data.

The elevation grid in the previous section is an example. Another example, below, is from the wireless telecommunications industry.

The yellow, orange and red colours in the grid represent the strength of a wireless signal. This is a numeric grid expressed in decibels.

wireless telecom
Note that Pitney Bowes products do not generate these radio frequency (RF) coverage maps but our products can work with them. Our partner, InfoVista, specialise in producing solutions for the industry to design and build wireless networks. Their solutions are built on a base of MapInfo technology.

Classified grids

In a classified grid, each cell is given a discrete value. The discrete values could be numeric or text but if numbers are used they do not necessarily imply an order or progression. This type of grid often represents data that has been previously classified, such as boundaries with demographic information or land use/zoning classifications. A common example is terrain classifications. Each grid cell is classified into one of a number of different types of terrain such as urban, industrial, forest, scrub, water, mountainous, etc. This is sometimes called a land use / land clutter grid.

classified grid
An example of a classified grid. This is called a land use/land clutter grid. One use for this is in wireless telecommunications companies. It is used in combination with elevation data in the algorithms calculating radio wave propagation, and it is also used when comparing signal strength with areas where people live, work and commute.

Example of using vector and raster data together

Let’s say you have a table of houses and you wish to know the elevation of each of the houses. You also have an elevation grid.

Using a raster analysis tool, including MapInfo Pro Advanced, Discover 3D, Engage 3D or Vertical Mapper, it is possible to assign the elevation of the underlying grid to the points.

point inspection
In the map above, the elevation value from the underlying grid has been assigned to each of the points. The points have been labelled with the elevation value. This capability is referred to as “point inspection” in our raster/grid add-on products.

Create grids from your own data through interpolation and modelling

One of the core capabilities of any raster based GIS is to create a representation of point data as a continuous grid. For example, points of elevation data can be used to build a surface representation. Likewise, a set of incidents can be used to create a hotspot density map.

classified grid

There are different mathematical methods for computing the grid representation from the point data. Above are just two examples, Inverse Distance Weighting and Hotspot Density.

Multi-spectral and hyper-spectral imagery

Imagery data can be collected and viewed in different wavelengths. Much of what you are used to viewing is imagery in the visible light spectrum. With the right data collection process and the right software it is possible to work with data from other wavelengths such as infrared. It is also possible to selectively view the red, green and blue light channels to bring out different detail in your data.

The screen shot below shows four examples from a Landsat 8 multi-spectral image.

classified grid

Landsat 8 data is a global source of imagery data available for free download. Click here for an article with more details. http://web.pb.com/mapinfopro-dec-2015/Download-Landsat-8-Data/

Want to try MapInfo Pro Advanced?

The free trial of MapInfo Pro v15.2 includes the MapInfo Pro Advanced capabilities. Click here to download the trial. If you already have our latest 64 bit release installed and you wish to try MapInfo Pro Advanced, go to the Licensing tab in the Backstage area. There you will find an option to start a 30 day trial of MapInfo Pro Advanced.

Article by Tom Probert, Editor of "The MapInfo Professional" journal

When not writing articles for "The MapInfo Professional", Tom enjoys talking to MapInfo Pro users at conferences and events. When not working he likes to see movies with car chases, explosions and kung-fu fighting.