- MIVIS deinstallation from the airborne platform 
and repetition of test bench scanner calibration to 
check the MIVIS performance level after the 
deployment. 
3. HYPERSPECTRAL DATA PRODUCTION 
The production (preprocessing and analysis) of these 
large hyperspectral data sets is performed in the 
LARA Project headquarter of Pomezia,a small town 
close to Rome, where a computer system with a 
software designed to handle MIVIS data has been 
established. 
The MIDAS production software (MIDAS: 
Multispectral Interactive Data Analysis System), 
besides the data production management, offers a 
powerful and highly extensible hyperspectral 
analysis capability. The Pomezia ground station is 
designed to maintain and check the MIVIS 
performance through the evaluation of data quality 
(like spectral accuracy, signal to noise performance, 
signal variations, etc.), and to produce, archive, and 
diffuse MIVIS data in the form of geometrically and 
radiometrically corrected data sets on low cost and 
easy access CC media. 
3.1 Avaiable products 
is structured as 
The MIDAS production pipeline 
follows: 
-MIVIS data ingestion into MIDAS system for 
extraction of all ancillary info and backups. 
-Production of 512 lines scenes in four different 
levels of preprocessing, as function of data owners 
request. The 4 different preprocessing levels are: 
- Level 0: Raw data + ancillary info for radiometric 
calibration and geometric corrections. 
- Level 1: Data radiometrically calibrated at the 
sensor, by using MIVIS internal references and the 
scanner's sensitivity measurements recorded from 
test bench. 
- Level 2: Data corrected by atmospheric effects and 
radiometrically corrected as in Level 1. In this case 
data owners are requested to supply all the paameters 
for path radiance and solar irradiance corrections. 
- Level 3: Level 1 or 2 data geometrically corrected 
from panorama distortions and the effects due to roll, 
pitch and yaw platform variations. The corrections 
are performed by using info gathered by MIVIS PAS 
system. 
- Level 4: Level 1 or 2 data geocoded images obtained 
integrating with a digital elevation model the MIVIS 
ancillary position and navigation sensors, allowing 
an exact positioning of pixels on the ground in no 
flat areas. 
Distribution of preprocessed data by using exabyte 
8mm tapes in tar UNIX format as support. 
131 
4. MIVIS CAMPAIGNS 
MIVIS data were recorded over different types of sites 
A large area, near the town of Trecate (Northern 
Italy), has been interested by a huge oil blow-out 
from oil rig belonging to AGIP (Italian Petroleum 
Agency). 
MIVIS has flown at an altitude of about 2000 m a.s.l. 
offering good quality data for the estimation of the 
level of oil pollution over cultivated and urban areas. 
MIVIS data over Garda Lake, Venice and Orbetello 
Lagoon and Po river delta, gave the possibility of a 
preliminary new insight in the field of specific 
applications of hyperspectral remote sensing to 
water quality, shallow and algal bloom analysis. 
MIVIS survey was carried out in Sicily for 
geological mapping and to investigate active 
volcanic phenomena (Mt. Etna and Eolian Islands), 
coastline pullution (Gela, Acireale, Taormina), ocean 
currents dynamics and algae bloom (Messina Straits 
and Marsala lagoon), beech-wood canopy (Mt. Etna 
slopes), waste discharges (Acireale), and 
archeological sites (Selinunte, Alesa and Acireale). 
In the summer of 1995 MIVIS has been deployed over 
the whole Italian Molise region for landslides and 
vegetational studies, in test sites located in Germany 
for enviromental investigations,and in Southern 
France for agricoltural studies. 
In the fall of 1995 a further deployment interested the 
Campania Region for the study of several specific 
targets, i.e: urban areas (Napoli), vulcanology 
(Vesuvius and Solfatara in Pozzuoli) coastal 
pollution, archeology (Pompei) and other 
enviromental and monumental sites (Caserta Royal 
palace and gardens. 
Two year of activity has shown as CNR/LARA 
Project has been and is operative according to the 
guidelines that represented the main tasks for the 
Project institution. The flight and data requests from 
the scientific national and international community 
has been far more above the initial forecast, and Lara 
Project is still working to improve its own 
capability to assure the most opportune reaction 
towards the future flight requests and shortest time for 
data production and diffusion amongst the users. 
Reference 
Bianchi R., Ferroni P., Marino C.M., Pignatti S., 
and Sangiorgio D 1994. CNR LARA Project Italy: 
multispectral interactive data analysis system. 
Proceedings of the first International Airborne 
Remote Sensing Conference and Exihibition” 
Strasbourg, V. III. 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B1. Vienna 1996