Happy new year and apologies for cross-posting...
I have just started work on a project for the Department for International
Development (DFID) and CASA and the University of Westminster on a project
looking at affordable mapping for developing nations.
The idea of the project is to address some of the problems of unregulated
urbanisation which can't be measured currently as there is almost no
national mapping programs in some countries, and no new maps created since,
in some cases, the 1960s.
The plan at first is to write a review of what projects have been started
in the past, plus the key techniques and references. I envisage that the
practical side will be in the area of:-
1. Automatic feature extraction to give us a good idea of the shape / size
of the cities
2. Population estimation based on lights source (i suppose) to create
thematic maps of the habitations.
I was wandering whether any of you could point me in the direction of data,
techniques and the key works in this field.
I have many resources on automatic feature extraction,edge detection and
the like. I also have access to agent-based modelling, CA algorithms and
the like, specifically for the growth of favella / shanty towns / peri-
What i also don't have is a guide to where i can get free (or very, very
cheap - it is for the developing world after all!) satellite imagery. Or do
we have to pay for everything? I expected old data (if you haven't got any
new maps since the 1960s, then being 5 or 10 years old isn't that much of a
problem) to be freely available and to only have to pay for the new stuff.
However, after a quick trawl around the internet I can't find anything...
I am also a bit unsure about the population estimation thing.
also, did someone mention Mandelbrot and a fractal algorithm for modelling
I know this is a long message but any help you can give me will be greatly
Centre for Advanced Spatial Analysis
University College London
1-19 Torrington Place
London WC1E 6BT
Tel. +44 (0)20 7679 4264
Fax.+44 (0)20 7813 2843
Mobile: 0798 520 6591
Email. [log in to unmask]