France. September 1-3, 2010 
In: Paparoditis N., Pierrot-Deseilligny M.. Mallet C.. Tournaire O. (Eds). IAPRS. Vol. XXXVIII. Part ЗА - Saint-Mandé, France. September 1-3. 2010 
37 
'ariance of measurement 
ombination of the inverse 
r and the oscillation on a 
ther hand, the variance of 
aid approximate to the 
i to the diameter and the 
neter. 
TRAFFIC CLASSIFICATION AND SPEED ESTIMATION IN TIME SERIES OF 
AIRBORNE OPTICAL REMOTE SENSING IMAGES 
G. Palubinskas* and P. Reinartz 
effect of quantization in 
»urement accuracy of the 
)oth WCM and BCM. 
oises on the measurement 
German Aerospace Center DLR, 82234 Wessling, Germany - Gintautas.Palubinskas@dlr.de 
Commission 111 - WG 111/5 
IS 
KEY WORDS: Classification, estimation, modelling, change detection, sequences, aerial, optical, imagery 
)f Fiducials for Accurate 
, IEEE Transactions on 
gence, 12(12), pp. 1196- 
ABSTRACT 
Dpment of Hybrid Video 
) Image Sensing, Journal 
my and Remote Sensing, 
Y, Hasegawa, H., Imoto, 
rithout Control Points by 
IEICE Transactions on 
pp. 1391-1400. 
Harley, I., 2006. Close 
tblishing, Caithness, pp. 
u, D. A., 1993. Particle 
Pail 1: Photogrammetric 
Experiments in Fluids, 15, 
okotsuka, H., Shirai, N.. 
thods of Circular Target 
of P hologramme tiy and 
In this paper we propose a new two level traffic parameter estimation approach based on traffic classification into three classes: free 
flow, congested and stopped traffic in image time series of airborne optical remote sensing data. The proposed method is based on 
the combination of various techniques: change detection in two images, image processing such as binarization and filtering and 
incorporation of a priori information such as road network, information about vehicles and roads and finally usage of traffic models. 
The change detection in tw'o images with a short time lag of several seconds is implemented using the multivariate alteration 
detection method resulting in a change image w'here the moving vehicles on the roads are highlighted. Further, image processing 
techniques are applied to derive the vehicle density in the binarized and denoised change image. Finally, this estimated vehicle 
density is related to the vehicle density, acquired by modelling the traffic flow for a road segment. The model is derived from traffic 
classification, a priori information about the vehicle sizes and road parameters, the road network and the spacing between the 
vehicles. Then, the modelled vehicle density is directly related to the average vehicle speed on the road segment and thus the 
information about the traffic situation can be derived. To confirm our idea and to validate the method several flight campaigns with 
the DLR airborne experimental wide angle optical 3K digital camera system operated on a Do-228 aircraft were conducted. 
Experiments are carried out to analyse the performance of the proposed traffic parameter estimation method for highways and main 
streets in the cities. The estimated speed profiles coincide qualitatively and quantitatively well with the reference measurements. 
1. INTRODUCTION 
During the past years, increasing traffic appears to be one of the 
major problems in urban and sub-urban areas. Traffic 
congestion and jams are one of the main reasons for immensely 
increasing transportation costs due to the wasted time and extra 
fuel. Conventional stationary ground measurement systems 
such as inductive loops, radar sensors or terrestrial cameras are 
able to deliver precise local traffic data with high temporal 
resolution, but their spatial distribution is still limited to 
selected motorways or main roads. 
A new type of information is needed for a more efficient use of 
road networks. Remote sensing sensors installed on aircrafts or 
satellites enable data collection on a large scale thus allowing 
wide-area traffic monitoring. Synthetic aperture radar (SAR) 
sensors due to their all-weather capabilities seem to be well 
suited for such type of applications. Ground moving target 
indication approaches based on the Displaced Phase Center 
Arrays technique are currently under investigation for airborne 
SAR sensors and space borne satellites, e.g. TerraSAR-X, but 
still suffer from the low vehicle detection rate, quite often 
below 30% (Meyer 2007). Traffic monitoring from optical 
satellites is still limited due to the not sufficiently high spatial 
resolution, but the detection of vehicle queues seems to be 
promising (Leitloff 2006). As it is shown already in (Reinartz 
2006, Hinz 2008) airborne optical remote sensing technology 
has a great potential in traffic monitoring applications. Several 
airborne optical remote sensing systems are already in 
experimental use at the German Aerospace Center DLR, e.g. 
airborne 3K camera system, consisting of three digital cameras 
capable of acquiring three images per second (Kurz 2007). and 
LUMOS (Ernst 2003). Automatic detection of vehicles and 
estimation of their speeds in sequences of optical images is still 
a challenge. Most known approaches are image based and still 
result in a too low completeness (e.g. less than 70%) thus being 
not yet suitable e.g. for the estimating of the traffic density and 
flow (Rosenbaum 2008). 
In this paper we propose a new model based approach and 
investigate its potential for the traffic parameter estimation in 
congested situations in sequences of airborne optical remote 
sensing data. Instead of detecting each individual vehicle and 
then estimating its speed (microscopic model) as e.g. in 
(Rosenbaum 2008) we exploit a linear vehicle density-speed 
relationship for a road segment (macroscopic model) to derive 
vehicle speeds from the estimated vehicle densities in an image. 
* Corresponding author.