ul 2004 
—— 
INTEGRATION OF AIRBORNE LASER DATA AND HIGH RESOLUTION 
SATELLITE IMAGES OVER LANDSLIDES RISK AREAS 
M. Barbarella, V. Lenzi, M. Zanni 
DISTART, University of Bologna, V.le Risorgimento 2, 40136 Bologna, Italy 
(maurizio.barbarella , veronica.lenzi , michela.zanni)@mail.ing.unibo.it 
KEY WORDS: LIDAR, radiometry, high resolution imagery, data integration, DTM, landslides 
ABSTRACT: 
When areas interested by catastrophic events like landslides are investigated, it is important to have a great amount of 
information with a greater than possible detail. 
It allows to evaluate the damages that occurred on the territory, to get a precise picture of post event situation and to understand 
the reasons and the effects of the event. 
This work starts from the availability of a great amount of different information (satellite images, airborne LIDAR data, DTM 
from maps, DTM from laser, ground surveys) over an area of Southern Italy that was involved in a significant landslide event in 
1998. Particularly, laser data have been used in order to improve the geometric and interpretative features of a satellite high 
resolution QuickBird image. In fact, laser data provide a double ‘product’: an elevation information, and, therefore, a Digital 
Terrain Model, and, at the same time, an intensity response. The DTM can be used to orthorectify the satellite images, the 
intensity response to create an intensity raster image as to increase the capability of analysis and interpretation of the images. 
In the present work the laser DTM accuracy has been evaluated in order to study the influence of different problems occurred 
during the laser survey, particularly problems with GPS signal acquisition. Furthermore, the work evaluates the accuracy of an 
orthoimage generated with a DTM obtained from the integration of height LIDAR data and a map derived DTM and analyses the 
use of intensity data to integrate the satellite image. 
1. INTRODUCTION area a great amount of data from different sources is available. 
À panchromatic QuickBird image, laser data and ground control 
  
Present work starts from the availability of a great amount of 
data over an area in the South of Italy (Sarno, Quindici and 
Bracigliano towns) hit in May 1998 by a catastrophic series of 
landslides that interested a wide and densely inhabited area. 
After the events, because of the lack of updated maps and data 
over the territory and in order to study the landslide phenomena, 
to prevent new disasters and to plan interventions and defensive 
works, different survey methodologies have been applied. 
Particularly, high resolution satellite images from different 
sensors (QuickBird and Ikonos) have been collected, ground 
surveys with traditional techniques (total station) and GPS over 
the landslide bodies and their surroundings and a LIDAR survey 
have been performed. 
The aim of the present work is to integrate and use different 
data in order to describe with the best accuracy the territory and 
the changement happened to the ground topography due to the 
event. Particularly a satellite image and laser data, both height 
and intensity, have been used. An orthoimage has been created 
using a Digital Elevation Model derived from the integration of 
à pre-event DTM and LIDAR DTM carried out over the most 
changed area. An evaluation of the accuracy achieved has been 
performed either using check points or trajectories obtained 
through kinematic GPS survey or comparing the landslide 
boundaries and other significant features achieved by the 
digitalisation performed over the intensity laser data. Moreover, 
an integration of image and intensity data has been performed in 
order to obtain a better description of the study area. 
2. STUDY AREA AND MATERIALS 
The study area is a mountainous area wide, approximately 75 
km”, with strong slopes, height up to 1000 m a.s.l. and almost 
completely covered by a dense vegetation. As said, over this 
points from GPS survey have been used in present work. 
The study area is represented in the white rectangle in figure |. 
1.1 High Resolution Image 
In order to frame the study area we have chosen to use a 
QuickBird image. It is a panchromatic image (BASIC) with a 
nominal spatial resolution of 0.6 m, collected over the entire 
area in January 2003. 
  
945