Wolfgang Herzner

Hierarchical Semantic Processing Architecture for Smart Sensors in Surveillance Networks

Data acquisition by multidomain data acquisition provides means for environment perception usable for detecting unusual and possibly dangerous situations. When being automated, this approach can simplify surveillance tasks required in, for example, airports or other security sensitive infrastructures. This paper describes a novel architecture for surveillance networks based on combining multimodal sensor information. Compared to previous methodologies using only video information, the proposed approach also uses audio data thus increasing its ability to obtain valuable information about the sensed environment. A hierarchical processing architecture for observation and surveillance systems is proposed, which recognizes a set of predefined behaviors and learns about normal behaviors. Deviations from “normality” are reported in a way understandable even for staff without special training. The processing architecture, including the physical sensor nodes, is called smart embedded network of sensing entities (SENSE).

High-level hierarchical semantic processing framework for smart sensor networks

This paper presents the framework of a novel approach to combine multi-modal sensor information from audio and video modalities to gain valuable supplementary information compared to traditional video-based observation systems or even just CCTV systems. A hierarchical, multi-modal sensor processing architecture for observation and surveillance systems is proposed. It recognizes a set of pre-defined behavior and learns about usual behavior. Deviations from ldquonormalityrdquo are reported in a way understandable even for staff without special training. The processing architecture including the physical sensor nodes is called SENSE (smart embedded network of sensing entities) (Zucker and Frangu, 2007).

Fault-based generation of test cases from UML-Models: approach and some experiences

In principle, automated test case generation - both from source code and models - is a fairly evolved technology, which is on the way to common use in industrial testing and quality assessment of safety-related, softwareintensive systems. However, common coverage measures such as branch or MC/DC1 for source code and states or transitions for state-based models provide only very limited information about the covered (implementation) faults. Fault-based test case generation tries to improve this situation by looking for detecting faults explicitly. This paper describes an approach combining fault- and model-based testing which has been realized in the European project MOGENTES2, using UML state machines for representing requirements, and discusses results of its application to a use case from the automotive domain.

CV-HAZOP: Introducing Test Data Validation for Computer Vision

Test data plays an important role in computer vision (CV) but is plagued by two questions: Which situations should be covered by the test data and have we tested enough to reach a conclusion? In this paper we propose a new solution answering these questions using a standard procedure devised by the safety community to validate complex systems: The Hazard and Operability Analysis (HAZOP). It is designed to systematically search and identify difficult, performance-decreasing situations and aspects. We introduce a generic CV model that creates the basis for the hazard analysis and, for the first time, apply an extensive HAZOP to the CV domain. The result is a publicly available checklist with more than 900 identified individual hazards. This checklist can be used to evaluate existing test datasets by quantifying the amount of covered hazards. We evaluate our approach by first analyzing and annotating the popular stereo vision test datasets Middlebury and KITTI. Second, we compare the performance of six popular stereo matching algorithms at the identified hazards from our checklist with their average performance and show, as expected, a clear negative influence of the hazards. The presented approach is a useful tool to evaluate and improve test datasets and creates a common basis for future dataset designs.

Encapsulating application subsystems using the DECOS core OS

This paper presents an approach to structured integration of different application subsystems on the same embedded hardware, as currently developed in DECOS (Dependable Embedded Components and Systems), an integrated project within the Sixth Framework Programme of the European Commission. Those application subsystems can have different criticality levels and vendors. Furthermore, reliable communication among application subsystems is a major concern. Focusing on the Encapsulated Execution Environment (EEE), which separates application subsystems in the space AND the time domain, this approach outlines the concepts and principles of an exokernel operating system, of partitioning, and of virtualization. The Core Operating System (COS) is described as a case study, including the hardware used, the current feature set, and benchmark values of central COS operations. This paper also presents a model for a platform-independent application interface layer. Parts of this interface layer are generated from task specification to provide tasks with tailored communication services.

Model-Based Development of Distributed Embedded Real-Time Systems with the DECOS Tool-Chain

The increasing complexity of distributed embedded systems, as found today in airplanes or cars, becomes more and more a critical cost-factor for their development. Model-based approaches have recently demonstrated their potential for both improving and accelerating (software) development processes. Therefore, in the project DECOS1, which aims at improving system architectures and development of distributed safety-critical embedded systems, an integrated, model-driven tool-chain is established, accompanying the system development process from design to deployment. This paper gives an overview of this tool-chain and outlines important design decisions and features. Copyright

Boltzmann Machine Topology Learning for Distributed Sensor Networks Using Loopy Belief Propagation Inference

Distributed sensor networks, as opposed to centralized networks, offer several advantages in terms of versatility and increased safety, which make their use particularly relevant for applications of security surveillance. A challenge of such systems is how to build autonomously a global description of the sensed environment without supervision of a central processing unit and with minimal configuration effort. We present an approach to ubiquitous computing, based on a semantic representation of the world view in terms of correlation of local information learned at the local level. There, a statistical description of the sensed activity is provided. Correlations of events among nodes are learned using a Boltzmann machine approach and used in order to establish neighborhood correspondences. Moreover, the communication between nodes is used to enrich the local description of the sensed environment by approximating the a-posterior distributions by marginal distributions computed with the loopy belief propagation algorithm. We present results of simulations emulating a security surveillance environment in which the sensors are cameras and activity is learned by processing video data.

Checking SCADE models for correct usage of physical units

Mismatches of units and of scales of values in physical calculations are disastrous, but rather common, in the development of embedded control systems. They can be as plain as mixing feet and metres, or as hidden as a wrong exponent in a complex calculation formula. These errors can be found by a checking algorithm, following some simple rules, if information on the units of the used variables is provided. This paper describes a developer friendly approach of providing this checking functionality in SCADE, a model-based graphical development tool for safety-critical embedded applications.

VITRO - Model based vision testing for robustness

This paper introduces a model-based approach for testing robustness of computer vision solutions with respect to a given task or application. Assessment of essential CV component robustness is crucial to ensure a safe robot and human coexistence. Currently this is mostly a manual and heuristic task lacking reliable metrics for determining the completeness and strength of a given test set. Our novel approach enables the generation of test data with a measurable coverage of optical situations both typical and critical for a given application. Typical situations are defined using a specific domain model while critical circumstances can be selected from a list of predefined hazards which was created using a proven hazard analysis procedure. Furthermore, the framework allows the automatic reduction of redundancy over the entire set of test images by using clustering. Finally the required oracle (ground truth) is automatically generated and is correct by definition.

Expressing Best Practices in (Risk) Analysis and Testing of Safety-Critical Systems Using Patterns

The continuing pervasion of our society with safety-critical cyber-physical systems not only demands for adequate (risk) analysis, testing and verification techniques, it also generates growing experience on their use, which can be considered as important as the tools themselves for their efficient use. This paper introduces workflow patterns to describe such best practices in a systematic way that efficiently represents this knowledge, and also provides a way to relate different patterns, making them easier to identify and use, and cover as wide a range of experiences as possible. The value of the approach is demonstrated using some pattern examples from a collection developed in the Artemis-project MBAT. Finally, the paper presents a wiki-based approach for developing and maintaining the pattern collection.