In: Wagner W., Székely, B. (eds.): ISPRS TC VII Symposium - 100 Years ISPRS, Vienna, Austria, July 5-7, 2010, IAPRS, Yol. XXXVIII, Part 7B 
415 
5.2.2. Estuaries 
5.2.3. Sea and ocean 
4.3. LAND COVER CLASSIFICATION USING 
REMOTE SENSING TECHNIQUES 
Everything on earth is changing with time. Land cover 
map can be a powerful tool to compare the changes of an 
area over time. It is impossible to cover a large area in short 
time through manual survey but with remote sensing (land 
cover map) it is an easier task. 
With land cover map it can be revealed how mach 
area of cities of Albania is using for what purpose, what are 
the pattern of land use change over time etc. which will 
help the policy makers to take necessary measure to ensure 
sound physical environment of the city etc. 
Land cover maps constitute necessary tools for 
development planning and management of the territory. 
Furthermore, land cover maps depicting the current reality 
are essential in countries where, due to political changes, 
rapid dynamic phenomena have taken place, resulting in a 
complete restructuring of the agricultural and other sectors, 
as in the case of Albania. 
For optimal use, land cover maps should be in digital 
format, which allows easy updating, and associated with a 
GIS including other information such as soil units, erosion 
features and provincial/municipal boundaries. The resulting 
database is an essential tool for decision-making in land 
management. 
But, before the creation of land cover map, it is 
necessary to have land cover classification (LCCS) (fig. 
Fig. 3.3.1. The place of LCCS in methodological approach 
of land cover maps creation 
Today, land cover classification based in use of 
satellite image data. The operational availability of high- 
resolution satellite imagery, namely Landsat TM, SPOT, 
Soyouz, ERS-SAR, RADARSAT and others, opens up new 
possibilities for investigating and monitoring natural 
resources. 
Satellite imagery is recorded in various 
wavelengths, visible and non-visible, which provide 
accurate information on ground conditions. Each object 
has unique and different characteristics of reflection or 
emission in different environment. An object and it’s 
environmental condition can be identified using reflected or 
emitted electromagnetic radiation from that object. 
The use of satellite image is popular world wide 
but its application is limited in Albania. However a land 
cover classification in our study is done. The classes were: 
1. Vegetation; 2. Built-up Area; 3. River/Deep Water; 4. 
Shallow Water; 5. Open Ground 
LANDSAT TM image was used for this study 
(table 3.3.1). 
able 3.3.1: 
Wavelength and a 
pplication of LANDSAT TM bands 
Band 
Wavelength 
(Pm) 
Application 
1 
0.45-0.52 
Coastal water mapping, soil 
vegetation differentiation, 
deciduous, coniferous 
differentiation 
2 
0.52- 0.60 
Green reflectance by healthy 
vegetation, excellent for 
pollution studies. 
3 
0.63-0.69 
Chlorophyll absorption for 
plant species differentiation 
identifies contrast 
between the vegetation 
classes. 
4 
0.76-0.90 
High reflectance for the 
vegetation, urban areas less 
reflective than the 
vegetation. Soil-crop and 
land-water contrasts are 
emphasized. 
5 
1.55-1.75 
Important for the crop 
identification, crop water 
content and soil moisture 
content. 
6 
10.4-12.5 
Hydro thermal mapping 
7 
2.08-2.35 
Plant heat stress 
The properties of the image were: 
• Image Sensor: LANDSAT TM 
• Image Format: BIL 
• Number of Lines: 2185 
• Number of Pixels per Line: 1441 
• Spatial Resolution: Band 1-5, 7 30m x 30m and band 6 
120mX120m 
• Spectral Resolution: 7 BANDS (1, 2, 3, 4, 5, 6, 7) 
Geocoding Wizard of ER Mapper was used for 
image registration and for registration, rectification and 
image classification the source image was converted from 
ERDAS LAN format to ER Mapper ERS format. The 
parameters of geocoding were: 
Geocoding type: Polynomial 
Polynomial order: Linear 
GCP picking method: A pre registered image of the 
area. 
The procedure of image enhancement makes it easy 
to identify and select ROIs (Region of Interest). In this 
study linear stretching and filter operations was performed 
for image enhancement.