Earth Observation Center

Knowledge of accurate travel times between various origins and destinations is a valuable information for daily commuters as well as for security related organizations (BOS) during emergencies, disasters, or big events. In this paper, we present a method for automatic estimation of travel times based on image series acquired from the recently developed optical wide angle frame sensor system (3K = "3-Kopf"), which consists of three non-metric off-the-shelf cameras (Canon EOS 1Ds Mark II, 16 MPixel). For the calculation of overall travel times, we sum up averaged travel times derived from individual vehicle velocities to pass defined road segments. The vehicle velocities are derived from vehicle positions in two consecutive geocoded images by calculating its distance covered over time elapsed. In this context, we present an automatic image analysis method to derive vehicle positions and vehicle distances involving knowledge based road detection algorithm followed by vehicle detection and vehicle tracking algorithms. For road detection, we combine an edge detector based on Deriche filters with information from a road database. The extracted edges combined with the road database information have been used for road surface masking. Within these masked segments, we extract vehicle edges to obtain small vehicle shapes and we select those lying on the road. For the vehicle tracking, we consider the detected vehicle positions and the movement direction from the road database which leads to many possible matching pairs on consecutive images. To find correct vehicle pairs, a matching in the frequency domain (phase correlation) is used and those pairs with the highest correlation are accepted. For the validation of the proposed methods, a flight and ground truth campaign along a 16 km motorway segment in the south of Munich was conducted in September 2006 during rush hour.

The usage of airborne camera systems for near real time applications will increase in the near future. This paper purposes a new hardware/software architecture to establish real time computation of images obtained from the DLR wide area airborne 3K-camera system. The main applications of our system are e.g. to monitor automotive traffic, to determine the workload of public road networks during mass events, or to obtain a survey of damages in disaster areas in real time. Therefore, many different image processing tasks have to be executed in real time. Orthorectification of images is necessary prior to all other processing tasks, e.g. before mapping data from street data bases into images or before tracking of vehicles. Nowadays, the calculation becomes possible due to fast graphic processing units (GPU) and with the support of a distributed real time system. In order to achieve real time image processing, we suggest a GPU-based algorithm for image orthorectification. The GPS/IMU-system provides the position and orientation of the aircraft with 128Hz quite accurately. Assuming synchronized measurements with the camera system and given camera calibration, direct orthorectification is implemented using OpenGL. Therewith, we are able to process high resolution images consisting of 16 MPixels with a frame rate of 3 Hz. This paper describes the implementation of the real time algorithm and gives first results.

Knowledge of accurate digital surface models (DSMs) is a valuable information for security authorities and organizations during emergencies, disasters, or big events. In this context, the potential of the recently developed digital optical camera system (3K) to derive DSMs automatically and in near real time will be evaluated. The 3K camera system with a frame rate of 3Hz consists of three non-metric off-the-shelf cameras (Canon EOS 1Ds Mark II, 16 MPixel), which are aligned in an array with one camera looking in nadir direction and two in oblique direction. The DSMs are calculated using a subpixel hierarchical matching based on interest points followed by a region growing algorithm. Exterior orientation is given online by GPS/IMU data, whereas interior camera parameters are provided in advance. The resulting DSM is calculated by forward intersection. For the validation, the point accuracies in relation to the base-to-height ratio as well as the reached point density in relation to the calculation time were calculated. Two flight campaigns with the 3K camera system were performed with varying frame rates over the centre of Munich on 30 th April and on 17 th June 2007. Point accuracies over flat terrain were validated with reference DEMs and correspond quite well to simulated accuracies. The validation showed that smaller base-to-height ratios decrease the point accuracies but increase point densities as images are more similar in particular over urban areas. Exemplarily possible applications for near real time DSMs in the context of disaster monitoring are described, e.g. monitoring of land slides, 3D change detection over urban area, and monitoring of infrastructure. Problematic in all cases is the long processing time, thus the matching algorithms must be modified or new techniques must be applied. Some starting points for the reduction of processing times are discussed. In this context, the proposed matching algorithms and results serve as reference for much faster implementations.

To estimate parameters of large-scale vegetation heterogeneity for the determination of soil heterogeneity a method based on fusion of multisensor image and pointwise ground-truth data is proposed. Regression analysis of multispectral imagery and ground-truth data is used to map soil indicative vegetation parameters for the area of a complete field. With highresolution aerial imagery natural and disturbances in a field, like tracks, weed-stripes and turning areas of crop machines, can be detected and excluded from mapping of vegetation heterogeneity. The disturbances caused by human cultivation and external natural influences hinder the estimation of the underlying soil condition. First results show the strong influence of disturbances on multispectral imagery and poor applicability of linear regression equations for predicting soil indicative vegetation parameters.

We propose a general framework to set up a method to estimate vegetation parameters from multidimensional remote sensing data using a physical model and a moderate amount of ground control data. This framework is exemplarily demonstrated for winter wheat fields imaged by the Daedalus multispectral scanner using the SAIL+PROSPECT model as physical model for leaves combined to a canopy layer. The vegetation parameters estimated with help of the devised method are intended to be used to derive information about soil heterogeneities which are important for precision farming.

We propose a general framework to set up a method to estimate vegetation parameters from multidimensional remote sensing data using a physical model and a moderate amount of ground control data. This framework is exemplarily demonstrated for winter wheat fields imaged by the Daedalus multispectral scanner using the SAIL+PROSPECT model as physical model for leaves combined to a canopy layer. The vegetation parameters estimated with help of the devised method are intended to be used to derive information about soil heterogeneities which are important for precision farming.

A near real time airborne monitoring system for monitoring of natural disasters, mass events, and large traffic disasters was developed in the last years at the German Aerospace Center (DLR). This system consists of an optical wide-angle camera system (3K system), a SAR sensor, an optical and microwave data downlink, an onboard processing unit and ground processing station with online data transmission to the DLR traffic and disaster portals. The development of the near real time processing chain from the data acquisition to the ground station is still a very challenging task. In this paper, an overview of all relevant parts of the airborne optical mapping system is given and selected system processes are addressed and described in more detail. The experiences made in the flight campaigns of the last years are summarized with focus on the image processing part, e.g. reached accuracies of georeferencing and status of the traffic processors.

In this paper we propose a new model based traffic parameter estimation approach in congested situations in time series of airborne optical remote sensing data. The proposed approach is based on the combination of various techniques: change detection, image processing and incorporation of a priori information such as road network, information about vehicles and roads and finally a traffic model. The change detection in two images with a short time lag of several seconds is implemented using the multivariate alteration detection method resulting in a change image where the moving vehicles on the roads are highlighted. Further, image processing techniques are applied to derive the vehicle density in the binarized 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 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 velocity 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 performed. Experiments are performed to analyse the performance of the proposed traffic parameter estimation method for highways and main streets in the cities. The estimated velocity profiles coincide qualitatively and quantitatively quite well with the reference measurements.

Purpose A new model based approach for the traffic congestion detection in time series of airborne optical digital camera images is proposed. Methods It is based on the estimation of the average vehicle speed on road segments. The method puts various techniques together: the vehicle detection on road segments by change detection between two images with a short time lag, the usage of a priori information such as road data base, vehicle sizes and road parameters and a simple linear traffic model based on the spacing between vehicles. Results The estimated speed profiles from experimental data acquired by an airborne optical sensor -3K camera system -coincide well with the reference measurements. Conclusions Experimental results show the great potential of the proposed method for the detection of traffic congestion on highways in along-track scenes.

Purpose Large area traffic monitoring with high spatial and temporal resolution is a challenge that cannot be served by today available static infrastructure. Therefore, we present an automatic near real-time traffic monitoring approach using data of an airborne digital camera system with a frame rate of up to 3 fps. Methods By performing direct georeferencing on the obtained aerial images with the use of GPS/IMU data we are able to conduct near real-time traffic data extraction. The traffic processor consists mainly of three steps which are road extraction supported by a priori knowledge of road axes obtained from a road database, vehicle detection by edge extraction, and vehicle tracking based on normalized cross correlation. Results Traffic data is obtained with a correctness of up to 79% at a completeness of 68%. Conclusions With this system we are able to perform areawide traffic monitoring with high actuality independent from any stationed infrastructure which makes the system well suited for deployments on demand in case of disasters and mass events.

We present an automatic traffic monitoring approach using data of an airborne wide angle camera system. This camera, namely the "3K-Camera", was recently developed at the German Aerospace Center (DLR). It has a coverage of 8 km perpendicular to the flight direction at a flight height of 3000 m with a resolution of 45 cm and is capable to take images at a frame rate of up to 3 fps. Based on georeferenced images obtained from this camera system, a near real-time processing chain containing road extraction, vehicle detection, and vehicle tracking was developed and tested. The road extraction algorithms handle a-priori information provided by a road database for a first guess of the location of the roads. Two different techniques can be used for road extraction. In the first method, roadside features are found by using an edge detector based on ISEF filtering, selecting the steepest edge, which is normally the edge between the tarry roads and the vegetation. The second method ...

Vehicle detection has been an important research field for years as there are a lot of valuable applications, ranging from support of traffic planners to real-time traffic management. Especially detection of cars in dense urban areas is of interest due to the high traffic volume and the limited space. In city areas many car-like objects (e.g., dormers) appear which might lead to confusion. Additionally, the inaccuracy of road databases supporting the extraction process has to be handled in a proper way. This paper describes an integrated real-time processing chain which utilizes multiple occurrence of objects in images. At least two subsequent images, data of exterior orientation, a global DEM, and a road database are used as input data. The segments of the road database are projected in the non-geocoded image using the corresponding height information from the global DEM. From amply masked road areas in both images a disparity map is calculated. This map is used to exclude elevated objects above a certain height (e.g., buildings and vegetation). Additionally, homogeneous areas are excluded by a fast region growing algorithm. Remaining parts of one input image are classified based on the 'Histogram of oriented Gradients (HoG)' features. The implemented approach has been verified using image sections from two different flights and manually extracted ground truth data from the inner city of Munich. The evaluation shows a quality of up to 70 percent.

A real time airborne monitoring system for monitoring of natural disasters, mass events, and large traffic disasters was developed in the last years at the German Aerospace Center (DLR). This system consists of an optical wide-angle camera system (3K system), a SAR sensor, an optical and microwave data downlink, an onboard processing unit and ground processing station with online data transmission to different end user portals. The development of the real time processing chain from the data acquisition to the ground station is still a very challenging task. In this paper, an overview of all relevant parts of the airborne optical mapping system is given and selected system processes are addressed and described in more detail. The experiences made in the flight campaigns of the last years are summarized with focus on the image processing part, e.g. reached accuracies of georeferencing and status of the traffic processors.

This paper examines and describes improvements of the processing time of an airborne wide area monitoring system which is under development at the German Aerospace Center (DLR). Aboard the aircraft a computer network equipped with three off-the-shelf cameras acquires images at intervals of up to 3fps. After orthorectification a traffic processor runs an automatic road traffic-data extraction and sends its result to a receiving ground station via an S-Band radio link. The processed results must be available as fast as possible. Hence, processing times of the computationally intensive tasks are meassured. In this context it turned out to consider not only the pure processing times but also the copy operations of the images between the modules. We compare several interprocess communication mechanisms and discuss the results. With the usage of relatively simple shared memory concepts a significant speed-up can be reached.

A world with growing individual traffic requires sufficient solutions for traffic monitoring and guidance. The actual ground based approaches for traffic data collection may be barely sufficient for everyday life, but they will fail in case of disasters and mass events. Therefore, a road traffic monitoring solution based on an airborne wide area camera system has been currently developed by DLR. Here, we present a new image processing chain for real-time traffic data extraction from high resolution aerial image sequences with automatic methods. This processing chain is applied in a computer network as part of an operational sensor system for traffic monitoring onboard a DLR aircraft. It is capable of processing aerial images obtained with a frame rate of up to 3 Hz. The footprint area of the three viewing directions of an image exposure with three cameras is 4 x 1 km at a resolution of 20 cm (recorded at a flight height of 1500 m). The processing chain consists of a module for data readout from the cameras and for the synchronization of the images with the GPS/IMU navigation data (used for direct georeferencing) and a module for orthorectification of the images. Traffic data is extracted by a further module based on a priori knowledge from a road database of the approximate location of road axes in the georeferenced and orthorectified images. Vehicle detection is performed by a combination of Adaboost using Haar-like features for pixel wise classification and subsequent clustering by Support Vector Machine based on a set of statistical features of the classified pixel. In order to obtain velocities, vehicle tracking is applied to consecutive images after performing vehicle detection on the first image of the burst. This is done by template matching along a search space aligned to road axes based on normalized cross correlation in RGB color space. With this processing chain we are able to obtain accurate traffic data with completeness and correctness both higher than 80 % at high actuality for varying and complex image scenes. The proposed processing chain is evaluated on a huge number of images including inner city scenes of Cologne and Munich, demonstrating the robustness of our work in operational use.

In this paper we analyze the usefulness of shadow to reconstruct height information from monocular images. Our goal is to derive height information from the measurement of the image coordinates of a point on a shadow edge and its conjugate on the object causing the shadow. We present the mathematical set-up of this approach and carry out an error analysis for the derived height. The proposed approach requires calibrated and oriented images and additionally information about the image capture time. Results are presented for synthetic and real data.

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.

Überwachungsaufgaben ist nun als Prototyp einsatzfähig. Der Sensor wurde gewichtsoptimiert, klein und mit preisgünstigen Bauteilen in einem Pylon seitlich am Hubschrauber konzipiert. Es mussten jedoch aufgrund der geforderten Funktionalität und der Echtzeitfähigkeit real-time Fähigkeit hier Kompromisse eingegangen werden. Integriert sind neben der automatischen Orthophotoerstellung verschiedene thematische Prozessoren wie z.B. die automatische Verkehrsdatenextraktion. Der Sensor ist mit der neuesten Generation handelsüblicher High-End Kleinbildkameras bestückt, die je nach Anwendung konfiguriert werden können. Dabei decken die Kameras bestückt mit 50mm Objektiven ein Blickfeld von bis zu 104° ab, beim Einsatz von 100mm Objektiven werden Auflösungen bis zu 3.5 cm bei einer Flughöhe von 500m ü.G. erreicht. Zusätzlich kann mit einer Kamera der nahe Infrarotbereich erfasst werden oder ein hochaufgelöstes Video (4k Video) aufgenommen werden. In diesem Paper werden die Systemkomponenten, die technischen Eigenschaften des Sensors beschrieben sowie die verschiedenen Einsatzmöglichkeiten des Systems skizziert.

Remote sensing applications like disaster or mass event monitoring need the acquired data and extracted information within a very short time span. Airborne sensors can acquire the data quickly and on-board processing combined with data downlink is the fastest possibility to achieve this requirement. For this purpose, a new low-cost airborne frame camera system has been developed at the German Aerospace Center (DLR) named 3K-camera. The pixel size and swath width range between 15 cm to 50 cm and 2.5 km to 8 km respectively. Within two minutes an area of approximately 10 km x 8 km can be monitored. Image data are processed onboard on five computers using data from a real time GPS/IMU system including direct georeferencing. Due to high frequency image acquisition (3 images/second) the monitoring of moving objects like vehicles and people is performed allowing wide area detailed traffic monitoring.