I guess it shouldnt be a problem, recorgizing slums. The structural difference, size and lack of proper road network should be easily picked up. If you mean to identify by automatic processing, use size and spahe feature.

A good start point would be the Global Report of Settlement 2003, UN-Habitat. It is called "The challenge of slums", and it contains one of the first operational definition of slum.

Hope it helps.

The identification of slums requires work in three dimensions i.e. legal, spatial and social. Tax/Revenue department can help to deal with legal dimension; spatial dimension can be dealt with RS/GIS datasets for indentifying infrastructural facilities/ connectivity (here I fully agree with Saif Uddin). To meet with social dimension, local groups/communities may be approached to get their feedback about slums in a certain area/territory. Therefore, identification of slums is not 100% automated process. It requires empirical study as well.

In India, through the Rajiv Awas Yojana, several cities have created a list of slums and conducted a household survey of these slums. Some cities even created GIS data base for slums. Some of the data is available online, look for Ministry for Poverty Alleviation - Rajiv Awas Yojana.

Not only in India, almost every country has a standard definition of either slum area or slum house (s).

A link to a paper that may help.

B.W. Wekesa a,b,*, G.S. Steyn a,1, F.A.O. (Fred) Otieno

A review of physical and socio-economic characteristics and intervention

approaches of informal settlements.

Habitat International 35 (2011) 238-245

doi:10.1016/j.habitatint.2010.09.006

In Rome too, in 1958 were defined slums in census

Dear Salvucci I'm interested to census data in Italy where were defined slums, if it is possible. Thank you

In reality nowadays in census data we don't speak of slum but we count under the voice other types of dwelling a similar situation.

The identification of slums are based on the definition your country adopts in defining the slums. Based on the physical features, you can use the digital image (based on roof type, building density and location) which can be used to mark the cluster or habitation as slums. You can detect the legal slums verses the new or illegal slum using the time series data (two time period images). Census if it identifies the housing quality and access to services for households can be used for demarcating or listing these areas as slum. The socio-economic data where poorest of the poor lives can also be one of leading data for locating the slums. But in all the cases poor do not live in bad quality shelters. The best is to conduct a physical site surveys of prospective areas identified by the images to finalize and listing as the slums. Every city has varied set of slums defined by location, community, income group and housing typologies. Next step, distinct slums based on classification. Ultimately, simple mapping is one exercise but if you need to have public policy, you need to collect more specific information.

I'm agree with Kumar. In Italy you could talk about slums as more suburban neighborhoods characterized by a concentration of illegal welfare states also including low and precarious houses and not alone. It's very different from slums in India or favelas in Brazil

at local level, local planning agencies and development authorities are responsible for identifying slum areas. Meanwhile, every dilapidated structure can be called a slum. Residential neighborhoods without amenities are also called slums. Thanks

Dear all, about this topic, someone is interested to write a paper with me?

my email [email protected]

in Brazil, we have data from the Brazilian Institute of Geography and Statistics. They have information on Census (1990, 2000, 2010). Spatial and statistic data that can be free downloaded on their website. Also, most brazilian municipalities has developed their Social Housing Plans and, usually, they have a better and more accurate data available..

Very interesting discussion. Donatella, any particular area you are focusing on with respect to slums?

thank you all, i really appreciate.

Thank you all. I finished to write a paper "Evaluating the use of the Web for slum tourism" (presented to Destination slum Conference, Potsdam, may 2014). This paper evaluates tour operator websites to analyse different types of strategic choices.  The web sites studied were of South Africa, Brazil and India areas.  

Could be interesting to define and compare slum for different countries.. 

Hi Donatella, if this paper is something you can share I'd be very interested in reading it. Concerning the definitions for slums, the state of the world cities report has various examples for slum definitions for different countries. Of course, this does not cover all slums and all countries.  

Identified from the air? It will be a lot of (relatively small pieces of) metal roofing.