Given that ground stationary infrastructures for traffic monitoring are barely able to handle everyday traffic volumes, there is a risk that they could fail altogether in situations arising from mass events or disasters. In this work, we present an alternative approach for traffic monitoring during disaster and mass events, which is based on an airborne optical sensor system. With this system, optical image sequences are automatically examined on board an aircraft to estimate road traffic information, such as vehicle positions, velocities and driving directions. The traffic information, estimated in real time on board, is immediately downlinked to a ground station. The airborne sensor system consists of a three-head camera system, a real-time-capable GPS/INS unit, five industrial PCs and a downlink unit. The processing chain for automatic extraction of traffic information contains modules for the synchronization of image and navigation data streams, orthorectification and vehicle detection and tracking modules. The vehicle detector is based on a combination of AdaBoost and support vector machine classifiers. Vehicle tracking relies on shape-based matching operators. The processing chain is evaluated on a large number of image sequences recorded during several campaigns, and the data quality is compared to that obtained from induction loops. In summary, we can conclude that the achieved overall quality of the traffic data extracted by the airborne system is in the range of 68% and 81%. Thus, it is comparable to data obtained from stationary ground sensor networks.

Automatic crowd detection in aerial images is certainly a useful source of information to prevent crowd disasters in large complex scenarios of mass events. A number of publications employ regression-based methods for crowd counting and crowd density estimation. However, these methods work only when a correct manual count is available to serve as a reference. Therefore, it is the objective of this paper to detect high-density crowds in aerial images, where counting– or regression–based approaches would fail. We compare two texture–classification methodologies on a dataset of aerial image patches which are grouped into ranges of different crowd density. These methodologies are: (1) a Bag–of–words (BoW) model with two alternative local features encoded as Improved Fisher Vectors and (2) features based on a Gabor filter bank. Our results show that a classifier using either BoW or Gabor features can detect crowded image regions with 97% classification accuracy. In our tests of four classes of different crowd-density ranges, BoW–based features have a 5%–12% better accuracy than Gabor.

This paper proposes a new method for estimating human crowd densities from aerial imagery. Applications benefiting from an accurate crowd monitoring system are mainly found in the security sector. Normally crowd density estimation is done through in-situ camera systems mounted on high locations although this is not appropriate in case of very large crowds with thousands of people. Using airborne camera systems in these scenarios is a new research topic. Our method uses a preliminary filtering of the whole image space by suitable and fast interest point detection resulting in a number of image regions, possibly containing human crowds. Validation of these candidates is done by transforming the corresponding image patches into a low-dimensional and discriminative feature space and classifying the results using a support vector machine (SVM). The feature space is spanned by texture features computed by applying a Gabor filter bank with varying scale and orientation to the image patches. For evaluation, we use 5 different image datasets acquired by the 3K+ aerial camera system of the German Aerospace Center during real mass events like concerts or football games. To evaluate the robustness and generality of our method, these datasets are taken from different flight heights between 800 m and 1500 m above ground (keeping a fixed focal length) and varying daylight and shadow conditions. The results of our crowd density estimation are evaluated against a reference data set obtained by manually labeling tens of thousands individual persons in the corresponding datasets and show that our method is able to estimate human crowd densities in challenging realistic scenarios.

Real-time monitoring of natural disasters, mass events, and large accidents with airborne optical sensors is an ongoing topic in research and development. Airborne monitoring is used as a complemental data source with the advantage of flexible data acquisition and higher spatial resolution compared to optical satellite data. In cases of disasters or mass events, optical high resolution image data received directly after acquisition are highly welcomed by security related organizations like police and rescue forces. Low-cost optical camera systems are suitable for real-time applications as the accuracy requirements can be lowered in return for faster processing times. In this paper, the performance of low-cost camera systems for real-time mapping applications is exemplarily evaluated based on already existing sensor systems operated at German Aerospace Center (DLR). Focus lies next to the geometrical and radiometric performance on the real time processing chain which includes image processors, thematic processors for automatic traffic extraction and automatic person tracking, data downlink to the ground station, and further processing and distribution on the ground. Finally, a concept for a national airborne rapid mapping service based on the low-cost hardware is proposed.

In this paper, we propose a navigation approach for smartphones that enables visitors of major events to avoid crowded areas or narrow streets and to navigate out of dense crowds quickly. Two types of sensor data are integrated. Real-time optical images acquired and transmitted by an airborne camera system are used to compute an estimation of a crowd density map. For this purpose, a patch-based approach with a Gabor filter bank for texture classification in combination with an interest point detector and a smoothing function is applied. Furthermore, the crowd density is estimated based on location and movement speed of in situ smartphone measurements. This information allows for the enhancement of the overall crowd density layer. The composed density information is input to a least-cost routing workflow. Two possible use cases are presented, namely (i) an emergency application and (ii) a basic routing application. A prototypical implementation of the system is conducted as proof of concept. Our approach is capable of increasing the security level for major events. Visitors are able to avoid dense crowds by routing around them, while security and rescue forces are able to find the fastest way into the crowd.