This paper describes the production of a particular type of density information 
from spatial data for spatial analysis applications or spatial data mining. The 
specification of an abstraction resolution for a chosen spatial framework is all
that is required to produce this density information. This detailed spatial 
framework used for the production of the density information should also be that
used for the subsequent spatial analysis. What is good about this way of 
producing density information is that once a spatial framework for the spatial 
analysis has been detailed, no further subjective bias is imposed on the 
resulting density information (even from NODATA space). In other words, the 
only spatial bias is given by the shapes of the frame used to tesselate data 
space. This paper focusses on the production of density information for two 
dimensional square grids or rasters containing NODATA space that are commonly 
use in geocomputation.

This paper describes the production of a particular type of density
information from spatial data for spatial analysis applications. 
The specification of an abstraction resolution for a chosen spatial framework
is all that is required to produce the density information. It is noted that
the specified abstraction resolution and spatial framework chosen for the
production of the density information should also be that for the subsequent
spatial analysis. What is good about this way of producing density information
is that once a spatial framework for the spatial analysis has been detailed, no
further subjective bias is imposed on the resulting density information. In
other words, the only spatial bias is given by the shapes used to tesselate
space. This paper focusses on the production of density information for two
dimensional square grids or rasters that are commonly use in geocomputation.



initial abstraction resolution of is all that is needed to drive the
production of density grids and their aggregation and cross scale
combination form the basis for an iterative process that is the focus of this paper.  This iterative process
converges like a smoothing algorithm looking about its neighbours, but it
converges more because of the limitations on the size and accuracy of numbers
that the computer can handle.    
In viewing computer generated and displayed maps there are several resolutions or generalisations that effect the image your brain interprets.  One 
relates to your own vision, that is, you can only see as well as your eyes let you.  Another is related to the resolution of the screen or other output 
device that you are using to look at the maps.  Another relates to the resolution of the software that I used to display and output the maps.  There are 
others, but perhaps the most important resolution or generalisation in terms of the differences between some of the maps in this directory that are 
density values of the same thing at the same scale relate to the abstraction of the initial grids that were aggregated to make these maps.  All the maps 
show the density of the roads over Great Britain (excluding some of the channel islands) as recorded in the Bartholomew's 1:200,000 scale (nominal) 
data set that I downloaded via KINDS under the CHEST aggreement in March 2000 (albeit that the density is given at different resolutions and ).  
The data depicting roads that is available in this data are a set of line vectors that have been collected via various means and cartographically 
generalised to produce nice maps.  Clearly the road surface in reality is not a line on a flat area but it is a three dimensional surface on a three 
dimensional surface.  By ignoring the three dimensional nature of the land surface and treating the road as if it is on a level plain and disregarding the 
fact that the road surface is actually  it as flat and The Bartholomew's data  

I am interested in calculating the density of different features of the road network in space over time primarily for the analysis and explanation of 
road traffic accident patterns for my PhD.  I am also interested in abstraction generalisation aggregation disaggregation density scale and change over 
time as a spatial analytical scientific modelling type person.  I want to work out ways of working out what data is good for what and what things can 
usefully be done with data to render a particular model for a particular purpose.  Hopefully this page and some maps in this directory will provide me 
with a means to help in thidevelop these idea help me discuss The study of the density of road generally Below