Building types 
Type of project Shop- 
(Use for area house 
delineation) 
Detached 
house 
Semi-de 
tached 
row 
house 
walk-up 
apt. 
multi-st. 
apt. 
sites 
+ser- 
vices 
1. Private formal 
1.1 shophouse projects X 
1.2 Private housing proj. 
X 
X 
X 
X 
X 
1.3 Land subdivision proj. 
X 
1.4 Individual buildings X 
X 
X 
2. Public formal 
2.1 NHA projects X 
X 
X 
X 
X 
X 
X 
2.2 Non NHA projects 
X 
X 
X 
3. Informal sector 
3.1 Slum/Squatter types 
X 
X 
3.2 Potential slums 
X 
X 
3.4 Mini slums 
X 
3.5 Klong houses 
X 
6.5 Interpretation criteria for residential class 
delineation: 
Following criteria have been used for interpreting 
the residential areas: 
Class Criteria used. 
from the airphoto to compute the number of units. 
According to the local building code, the number of 
housing units in one structure will never exceed 24 
units, which implies that sometimes a wider frontage 
from 5 up to 8 meter must be applied. Data on hous 
ing units in multi-storey buildings were preferably 
obtained from fieldchecks. 
Shophouse 
Private housing 
project 
Land subdivision 
Individual project 
Public sector 
Informal sector 
Shape of building, relative 
locations and stereo height. 
Similarity of housing type, 
repetition of road pattern 
relative location (mostly newly 
developed area) and mostly no 
vacant plot in between each 
house. 
Similarity in housing lay-out, 
repetition of road patterns, 
low density because a number of 
vacant plots exist in between 
each house, relative location 
(mostly newly developed area). 
Areas which do not belong to the 
first 3 afore mentioned classes, 
and small scale. 
Not interpreted, data from other 
sources. 
Size of each house relatively 
small, irregular pattern of lay 
out and settlement boundary with 
high density, relative location 
(behind shophouse or proximity 
to industrial land use, or along 
public land such as railway or 
canal area where general 
condition is worse, image is 
shown in texture). 
6.7 Housing unit estimate in the informal settlement 
In the informal settlement area, the landuse 
method as developed by Krause et al (1974) has been 
applied. Types of informal settlement were clas 
sified also according to their densities: 
Type of settlement 
Density applied 
1. Squatter type 
2. Slum type 
3. Potential slum 
4. Potential slum 
class 1 dg = 
class 2 d^ = 
150 houses/hectare 
130 houses/hectare 
100 houses/hectare 
80 houses/hectare 
The measurement of area per type of settlement was 
done by counting dots with a ’dot grid*, and as 
signed a code i.e. A^.Ag.A^and A^. Total number of 
housing stock in each settlement type were derived 
from: 
H = A. d 
where H = total housing unit 
A = total area devoted to each type of 
settlement 
d = the average housing density belong 
to that type of settlement. 
To help the photo interpreters with the inter 
pretation criteria, a stereoscopic photo-key was 
made for reference by the interpreters. Fig.l il 
lustrates some of the main elements from this photo 
key. 
6.6 Counting of houses 
The value for each housing density (d) was derived 
from the average number of houses counted in a 
hectare block of each particular type of settlement. 
The areas for counting that type of settlement were 
selected prior to the available updated large scale 
map so that comparative counting from airphoto and 
map of the same area could be done. 
The housing unit count depended to some extent on 
the local frame of reference. For certain types of 
housing, where individual units cannot be clearly 
seen on the airphoto, like rowhouse, shophouse or 
semi-detached house, some clues on environmental 
aspects, such as the presence of entrance, plot 
boundary on the front or back yard will be searched. 
In case no clues were identifiable, a standard 4 
meter front width was applied for shophouses and 
rowhouses. The length of the building was measured 
6.8 Final output 
By this methodology, data on number of housing 
stock and residential area by type of housing 
project and by type of housing of 1984 and 1974 were 
obtained in district wise. 
The total sum of all classes plus 5% is the total 
housing stock. This 5% forms the class ’other’ which 
housing units could not be identified from air