nd-
T.
ion
ion
110,
pp.
ntic
«IS
In:
ro
fa
In:
ro
sel
S: d.
vical
ation
and
s°99
DATA FUSION OF AERIAL IMAGES COLLECTED BY MATRIX CAMERA AND LINE SCANNER OF
DIFFERENT RESOLUTION
A. Krtalic*
ee ti tA PS
KEY WORDS: fusion, integration, multiresolution, multisensor, multispectral, resolution
ABSTRACT:
The information collected through the airborne remote sensing are useful for many scientific and practical works. In order to use
collected information and existing knowledge of the observed object in the most efficient way for the purpose of solving specific
problems, we have to examine their characteristics and if possible fuse them. This article is motivated by the existence of two
operative digital sensor systems in Croatia that collect information at different wavelengths and with different methods of taking
images. The first sensor is the digital matrix (staring) camera, for three visible channels (V:0.4-0.5 pm, 0.5-0.6 pm, 0.6-0.7 pm), for
near infrared channel (NIR: 0.7-1.0 um) and the second sensor is the longwave thermal infrared (TIR: 8-14 pm) parallel scan camera.
This paper will analyze the possibility of fusing information provided by above sensors, of one not very accessible area (mountainous
terrain, with large differences of heights). Digital sensor systems yield multispectral images with different spatial and radiometric
resolutions, geodetic and photogrammetric data (geographic maps, GPS data, ortophoto maps, geocoded mosaics), the insight in the
field situation (ground truth) and other accessible sources provide inputs for fusion. It will also consider influence of the fusion on the
results of classification of mosaics developed on multi-sensor, multi-spectral and multi-resolution digital images from above
mentioned two sensor systems. The mosaics made on the basis of the above-mentioned sensor systems have been separately made,
and later on the multi-spectral mosaics containing the information from all 5 channels will be created. Spatial and geometrical quality
of multi-sensor and multi-spectral images and mosaics were analysed in earlier works, so this article will not deal with this matter,
but will use the results of those earlier analyses. The work is an integral part of the research conducted in the scientific project ARC
funded by the European Commission.
1. INTRODUCTION
The operative system, which consists of two sensors, two
digital cameras, DuncanTech MS3100 and Thermovision
THV 1000, is in intesive operational use in Croatia for
remote sensing of minefields. The first sensor is the digital
matrix camera, for acquisition in three visible channels (V:
0.4-0.5 um, 0.5-0,6 pm, 0.6-0.7 pm) and near infrared (NIR:
0.7-1.0 pm) and the second sensor is the parallel scan camera
which collects data in the longwave thermal infrared (TIR: 8-
14 pm) area. These two cameras are different in many ways,
they collect data in different wavelengths and they have
different principle of collecting data. DuncanTech MS3100 is
the digital matrix camera (sensor resolution 1392x1040
pixels) and it produces images in central projection.
Thermovision THV 1000 collects the data by scaning 5
parallel lines at the time in 80 rows with resolution of sensor
5x400 pixels. The reason for this particular camera to be used
instead of matrix camera (available on the market) is the
sensor resolution. The matix TIR cameras have resolution of
320x240 pixels and for that reason the parallel scan camera is
in use. The flights for collecting the images for the purpose of
humanitarian demining in Croatia took place at the height of
130 m and higher above the ground. Because of small flight
height, the small surface of each image and a large number of
images, the orientation in space has become more difficult.
This problem can be solved by mosaicing the selected images
of the an area and geocoding of the whole mosaic. Therefore,
for practicle use of mosaics, it is necessary to know what can
we expect from it in geometric and radiometric way. In the
previous work (Krtali¢, A., Fiedler, T., 2003.) we considered
the spatial quality of both (TIR and VNIR) mosaics and in
this article the results of data fusion of data from this mosaics
and every other aviable information about one not very
accessible area (mined moutainous terrain, with large
differences of heights) was done and later on analized. This is
important especially for TIR mosaic (level in data fusion) and
its influence on multispectral image in order to improve beter
understanding of scene and for bether interpretation of the
terrain.
2. MOSAICING AND GEOCODING OF MOSAICS
For purpose of mosaicking there were 6 VNIR and 6 TIR
images chosen which were snapshots in two strips in the
same period of time. VNIR and TIR images covered
approximately the same surface of the terrain. VNIR images
are RGB images created from infrared (IR), red (R) and green
(G) channels. The correlation between blue and green
channel is over 9294, and for that reason the multispectral
(RGB) image was created with only 3 above mentioned
instead of all 4 chanells. The first step was mosaicking along
the strip and after that mosaiking between strips was carried
out. After mosaicking, the both mosaics were geocoded.
VNIR mosaic was geocoded according to digital orthophoto
(DOF) of the same area and later on the TIR mosaic
according to geocoded VNIR mosic. The size of pixels are:
0.5 m on DOF, calculated spatial resolution of VNIR mosaic
(according to flight height and width of used objective) is cca
0.2 m, and TIR mosaic cca 0.5 m. Geocoded mosaics, and
evry other outputs in this work have the same size of pixel
and that is 0.1 m. This action has been done so as to facilitate
the conversion of images of different resolution
(multiresolutions images) and different wavelenght
(multispectral) on the same size for farther integrations and
analysis (principal components, classification). After
geocoding all geodetic basis (map 1:5000, DOF 1:5000,
VNIR mosaik, TIR mosaik) was displayed on the screen one
above eachother (in layers), and cut in the same size. Now we
have multispectral images with the same resolution (but
different spatial resolution!) which cover the same terrain and
