Kristian Kroschel

A Robust and Efficient Approach for Human Tracking in Multi-camera Systems

In this paper, a robust and efficient approach for multicamera human tracking is presented. The approach is integrated in an experimental surveillance system, based on a camera network with a task-oriented architecture. At sensor level, image processing algorithms are applied for object detection and feature extraction. Additionally, for each object that is to be tracked, an agent-based multi-sensor process is created, which autonomously performs multi-sensor data association and fusion. One of the major challenges in such systems is to robustly determine correspondences between observations from different sensors with different environmental conditions. Therefore, in this paper, efficient and robust spacial and appearance features for object description and recognition are proposed. For spacial description an approximated object position in world coordinates is estimated and evaluated by an inconsistency detector before associated to a Kalman Filter. For appearance similarity calculation, an appearance model is proposed and a similarity metric based on the Earth Mover’s Distance (EMD) is presented. Finally, the data fusion algorithm based on these features for tracking objects in overlapping and non-overlapping camera networks is presented.

Kalman-Bucy-Filter

Wahrend im 4. Kapitel die auf den Arbeiten von Kalman beruhende Signalschatzung zeitdiskreter Prozesse betrachtet wurde, soll nun die Schatzung zeitkontinuierlicher Prozesse behandelt werden.

An Object- and Task-Oriented Architecture for Automated Video Surveillance in Distributed Sensor Networks

In this paper, an agent-based software architecture for automated wide area video surveillance systems is presented. The proposed concept is designed for detection and tracking of moving objects across multiple camera views. The surveillance system consists of a decentralized collaborative sensor network with object- and task-oriented architecture. At sensor node level, image processing algorithms are applied for event and object detection. In case of detection (e. g. motion) an agent-based multi-sensor processing cluster is created. Each instantiated cluster is responsible for observation of one object in the scene. Object handover is managed autonomously by the dynamic sensor clusters. The dynamic sensor clustering approach allows adding new sensors without resetting the system parameters, which is a big advantage in large sensor networks. Furthermore, by using the agent-based architecture it is possible to create a framework with an adaptive data and processing load. Additionally, upgrade of system capabilities can be done easily updating or adding new processing agents. The proposed concept has been proved on an experimental video surveillance system at the Fraunhofer IITB.

Assessment of image sensor performance with statistical perception performance analysis

The performance of perceptive systems depends on a large number of factors. The practical problem during development is, that this dependency is very often not explicitly known. In this contribution we address this problem and present an approach to evaluate perception performance, as a function of e.g. quality of the sensor data. The approach is to use standardized quality metrics for imaging sensors, and to relate them to the observed performance of the environment perception. During our experiments, several imaging setups were analyzed. The output of each setup is processed offline to track down performance differences with respect to the quality of sensor data. We show how and to what extend the measurement of the Modulation Transfer Function (MTF) using standardized tests can be applied to evaluate the performance of imaging systems. The influence of the MTF on the signal-to-noise ratio can be used to evaluate the performance on a recognition task. We assess the measured performance by processing the data of different, simultaneously recorded imaging setups for the task of lane recognition.

Auswertung von Vibrometersignalen zur Bestimmung kardiovaskulärer Parameter

Zusammenfassung Laser-Doppler-Vibrometer (LDV) messen kleinste Schwingungsamplituden und haben neben technischen auch biomedizinische Anwendungen wie zum Beispiel die Schwingungsmessung auf der Haut. Die Überwachung des Herzrhythmus wird aktuell mit dem Elektrokardiogramm (EKG) durchgeführt, wobei Elektroden auf der Haut befestigt werden. Eine berührungslose Messung wäre vorteilhaft und würde die Patientenüberwachung (z. B. verbrannte Haut, Traumata auf dem Brustkorb, Frühgeborene und Athleten unter Belastung) erheblich vereinfachen. In diesem Beitrag zeigen wir, dass optische Vibrokardiographie (VKG) mit einem LDV nicht nur wie bereits bekannt die Herzfrequenz und deren Variabilität messen kann, sondern auch eine berührungslose Herzrhythmusmessung mit einer robusten atrioventrikulärer (AV) Block-Klassifikation ermöglicht, da der Herzschlag als mechanische Quelle zu Schwingungen auf der Haut führt. Dazu haben wir einen geeigneten Bereich auf dem Thorax identifiziert, der Schwingungen aufweist, die eindeutig dem Schließen von der linken Herzkammer und der linken Herzvorkammer zugeordnet werden können. Wir diskutieren außerdem verschiedene Signalverarbeitungskonzepte für eine automatische Signalmustererkennung von charakteristischen Signalabschnitten.

Evaluating sensor effects on perception performance

Performance of perceptive systems depends on the quality of the input data. In this contribution, an approach to evaluate perception performance as a function of quality of the sensor data is presented. Standardized quality metrics support the imaging sensors performance measurement. Several imaging setups are analyzed with real world experiments. The output of each setup is processed offline to track down performance differences with respect to the quality of sensor data. An adapted measurement is calculated to measure the sensor performance with respect to the data quality for the involved perceptive components. The measured performance is assessed by processing the data of different simultaneously recorded imaging setups for the task of feature extraction of road lanes.

Automatic detection of measurement points for non-contact vibrometer-based diagnosis of cardiac arrhythmias

Monitoring of the heart rhythm is the cornerstone of the diagnosis of cardiac arrhythmias. It is done by means of electrocardiography which relies on electrodes attached to the skin of the patient. We present a new system approach based on the so-called vibrocardiogram that allows an automatic non-contact registration of the heart rhythm. Because of the contactless principle, the technique offers potential application advantages in medical fields like emergency medicine (burn patient) or premature baby care where adhesive electrodes are not easily applicable. A laser-based, mobile, contactless vibrometer for on-site diagnostics that works with the principle of laser Doppler vibrometry allows the acquisition of vital functions in form of a vibrocardiogram. Preliminary clinical studies at the Klinikum Karlsruhe have shown that the region around the carotid artery and the chest region are appropriate therefore. However, the challenge is to find a suitable measurement point in these parts of the body that differs from person to person due to e. g. physiological properties of the skin. Therefore, we propose a new Microsoft Kinect-based approach. When a suitable measurement area on the appropriate parts of the body are detected by processing the Kinect data, the vibrometer is automatically aligned on an initial location within this area. Then, vibrocardiograms on different locations within this area are successively acquired until a sufficient measuring quality is achieved. This optimal location is found by exploiting the autocorrelation function.

Ein auftragsorientierter Ansatz zur kameraübergreifenden Personenverfolgung in verteilten Kameranetzwerken

Zusammenfassung In diesem Artikel wird ein generischer Ansatz und eine Prozessarchitektur zur auftragsorientierten Sensordatenanalyse vorgestellt. Zunächst werden anhand von Anwendungsbeispielen aus der Videoüberwachung die Vorteile einer auftragsorientierten Informationsverarbeitung gegenüber dem klassischen sensororientierten Ansatz beschrieben. Des Weiteren wir eine generische Prozessarchitektur vorgestellt, welche eine effiziente auftragsorientierte Informationsverarbeitung ermöglicht. Schließlich werden spezielle Ausprägungen der eingeführten Systemkomponenten vorgestellt, um die exemplarische Anwendung der kameraübergreifenden Personenverfolgung zu realisieren. Abstract A generic approach and process architecture for task-oriented sensor data processing is presented. Hereby, we focus on the application of video surveillance in large camera networks. Firstly, advantages of the proposed task-oriented video processing, compared to today´s (sensor-oriented) video surveillance systems, are described. Secondly, a generic process framework is presented, which has been designed for task-oriented video processing. Finally, system components of the task-oriented framework needed for the task of multi-camera person tracking are introduced.

Performance prediction of vehicle detection algorithms

Performance prediction of computer vision algorithms is of increasing interest whenever robustness to illumination variations, shadows and different weather conditions has to be ensured. The statistical model which is presented in this contribution predicts the algorithm performance under the presence of noise, image clutter and perturbations and therefore provides an algorithm-specific measure of the underlying image quality. For the prediction of the detection performance logistic regression using covariates defined by the properties of the vehicle signatures is used. This approach provides an estimate of the probability of a single vehicle signature being detected by a given detection algorithm. To describe the relationship between background clutter and the false alarm rate of the algorithm a severity measure of the image background is presented. After the construction of the algorithm model, the probability of a vehicle signature being detected and the false alarm rate are estimated on new data. The model is evaluated and compared to the true algorithm performance.

Leistungsprädiktion von Algorithmen zur Mammographie-Bildauswertung

Der vorliegende Ansatz zeigt neue Methoden zur Vorhersage der Verfahrensleistung bei der Detektion von Mikrokalzifikationen in Mammogrammen auf. Dazu werden fur ein gegebenes Mammogramm-Auswertungsverfahren einerseits fur die zu detektierenden Mikrokalzifikationen, andererseits fur den verbleibenden Hintergrund-Bildbereich separate Modelle erstellt, aus denen verfahrensspezifische Bildqualitatsmase abgeleitet werden konnen. Um die Qualitat dieser Pradiktionsmodelle nachzuweisen, wird die Verfahrensleistung aufgrund der extrahierten Bildqualitatsmase geschatzt und mit der tatsachlich erzielten Verfahrensleistung verglichen.