Mathias Anneken
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Mathias Anneken has a total of 65 co-authors. Here are the top co-authors by number of shared publications:
- Manjunatha Veerappa
Fraunhofer IOSB , Karlsruhe , Germany
4 shared publications - Marco F. Huber
Fraunhofer Institute for Manufacturing Engineering and Automation IPA
4 shared publications Yvonne Fischer
Fraunhofer Institute of Optronics and System Technologies and Image Exploitation (Fraunhofer IOSB), Germany
4 shared publicationsJurgen Beyerer
Fraunhofer IOSB, Karlsruhe, Germany
3 shared publicationsJürgen Beyerer
Karlsruhe Institute of Technology (KIT), Fraunhofer Institute of Optronics and System Technologies and Image Exploitation (Fraunhofer IOSB), Germany
3 shared publications- Nadia Burkart
Fraunhofer IOSB
3 shared publications
Latest Publications
- 17 March 2023
- Published by Elsevier
- p. 173-189
Explaining CNN classifier using association rule mining methods on time-series
- 17 March 2023
- Published by Elsevier
- p. 173-189
Explaining CNN classifier using association rule mining methods on time-series
- 17 March 2023
- Published by Elsevier
List of contributors
- 22 September 2022
- Published by Walter de Gruyter GmbH in TM - Technisches Messen
- Vol. 90 (3) , 154-166
Explainable AI for sensor-based sorting systems
Explainable artificial intelligence (XAI) can make machine learning based systems more transparent. This additional transparency can enable the use of machine learning in many different domains. In our work, we show how XAI methods can be applied to an autoencoder for anomaly detection in a sensor-based sorting system. The setup of the sorting system consists of a vibrating feeder, a conveyor belt, a line-scan camera and an array of fast-switching pneumatic valves. It allows the separation of a material stream into two fractions, realizing a binary sorting task. The autoencoder tries to mimic the normal behavior of the nozzle array and thus can detect abnormal behavior. The XAI methods are used to explain the output of the autoencoder. As XAI methods global and local approaches are used, which means we receive explanations for both a single result and the whole autoencoder. Initial results for both approaches are shown, together with possible interpretations of these results.
- 22 September 2022
- Published by Walter de Gruyter GmbH in TM - Technisches Messen
- Vol. 90 (3) , 154-166
Explainable AI for sensor-based sorting systems
Explainable artificial intelligence (XAI) can make machine learning based systems more transparent. This additional transparency can enable the use of machine learning in many different domains. In our work, we show how XAI methods can be applied to an autoencoder for anomaly detection in a sensor-based sorting system. The setup of the sorting system consists of a vibrating feeder, a conveyor belt, a line-scan camera and an array of fast-switching pneumatic valves. It allows the separation of a material stream into two fractions, realizing a binary sorting task. The autoencoder tries to mimic the normal behavior of the nozzle array and thus can detect abnormal behavior. The XAI methods are used to explain the output of the autoencoder. As XAI methods global and local approaches are used, which means we receive explanations for both a single result and the whole autoencoder. Initial results for both approaches are shown, together with possible interpretations of these results.
- 1 February 2022
- Published by Elsevier in Journal of Computational Science
- Vol. 58
Validation of XAI explanations for multivariate time series classification in the maritime domain
- 1 February 2022
- Published by Elsevier in Journal of Computational Science
- Vol. 58
Validation of XAI explanations for multivariate time series classification in the maritime domain
- 1 December 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1048-1053
Synthetic Trajectory Extraction for Maritime Anomaly Detection
In the maritime domain, the main goal for large vessels is to drive as fast and efficiently as possible from one harbour to the next. This results in vessels following quite distinct routes. Deviations from these routes can be seen as indicators for illegal or at least suspicious behaviour. In order to extract these routes, clustering algorithms can be used, that will store the traffic patterns as synthetic trajectories. A vessel can be assigned to a specific route or a deviation from the routes can be detected by evaluating these route models. A challenge in designing the synthetic trajectories lies in the faithful representation, sparsity of data, treatment of outliers and split-up of routes. Here, a novel approach to model the uncertainty of the vessels movements while following a route is introduced: A cluster is now represented by a trajectory consisting of segments. Due to obstacles like shallow waters, a route might split up which will generate sub-segments. Instead of storing all cluster points, the position, speed, and course of the vessels are attached to each segment in form of normal distributions. Using these attributes, the probability of a deviation from the given trajectory can be estimated.
- 1 December 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1048-1053
Synthetic Trajectory Extraction for Maritime Anomaly Detection
In the maritime domain, the main goal for large vessels is to drive as fast and efficiently as possible from one harbour to the next. This results in vessels following quite distinct routes. Deviations from these routes can be seen as indicators for illegal or at least suspicious behaviour. In order to extract these routes, clustering algorithms can be used, that will store the traffic patterns as synthetic trajectories. A vessel can be assigned to a specific route or a deviation from the routes can be detected by evaluating these route models. A challenge in designing the synthetic trajectories lies in the faithful representation, sparsity of data, treatment of outliers and split-up of routes. Here, a novel approach to model the uncertainty of the vessels movements while following a route is introduced: A cluster is now represented by a trajectory consisting of segments. Due to obstacles like shallow waters, a route might split up which will generate sub-segments. Instead of storing all cluster points, the position, speed, and course of the vessels are attached to each segment in form of normal distributions. Using these attributes, the probability of a deviation from the given trajectory can be estimated.
- 1 June 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2015-2021
Body Pose and Context Information for Driver Secondary Task Detection
Distraction of the driver by secondary tasks is already dangerous while driving manually but especially in handover situations in an automated mode this can lead to critical situations. Currently, these tasks are not taken into account in most modern cars. We present a system that detects typical distracting secondary tasks in an efficient modular way. We first determine the body pose of the driver and afterwards use recurrent neuronal networks to estimate actions based on sequences of the captured body poses. Our system uses knowledge about the surroundings of the driver that is unique to the car environment. Our evaluation shows that this approach achieves better results than other state of the art systems for action recognition on our dataset.
- 1 June 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2015-2021
Body Pose and Context Information for Driver Secondary Task Detection
Distraction of the driver by secondary tasks is already dangerous while driving manually but especially in handover situations in an automated mode this can lead to critical situations. Currently, these tasks are not taken into account in most modern cars. We present a system that detects typical distracting secondary tasks in an efficient modular way. We first determine the body pose of the driver and afterwards use recurrent neuronal networks to estimate actions based on sequences of the captured body poses. Our system uses knowledge about the surroundings of the driver that is unique to the car environment. Our evaluation shows that this approach achieves better results than other state of the art systems for action recognition on our dataset.
- Open Access
- 1 January 2017
- Published by INSTICC
- p. 200-207
A Multi-agent Approach to Model and Analyze the Behavior of Vessels in the Maritime Domain
The automatic detection of suspicious behavior is one important part in order to support operators in surveillance tasks. Therefore, an approach to model the behavior of objects by using multi-agent systems is introduced. As each object has its own objectives and desires to fulfill, these are modeled as utility functions. The actions of the objects are estimated by using the Nash bargaining solution. Consequently, it is implied, that the objects are cooperating in order to achieve an optimal result for themselves. First results for this algorithm are shown by using examples from the maritime domain. On the one hand, the algorithm is used to calculate an anomaly score. On the other hand, it is used to predict the movement of vessels
- 1 January 2017
- Published by INSTICC
- p. 200-207
Open AccessA Multi-agent Approach to Model and Analyze the Behavior of Vessels in the Maritime Domain
The automatic detection of suspicious behavior is one important part in order to support operators in surveillance tasks. Therefore, an approach to model the behavior of objects by using multi-agent systems is introduced. As each object has its own objectives and desires to fulfill, these are modeled as utility functions. The actions of the objects are estimated by using the Nash bargaining solution. Consequently, it is implied, that the objects are cooperating in order to achieve an optimal result for themselves. First results for this algorithm are shown by using examples from the maritime domain. On the one hand, the algorithm is used to calculate an anomaly score. On the other hand, it is used to predict the movement of vessels
- 1 July 2016
- Published by Springer Nature
- p. 89-107
Quantitative Assessment of Anomaly Detection Algorithms in Annotated Datasets from the Maritime Domain
- 1 July 2016
- Published by Springer Nature
- p. 89-107
Quantitative Assessment of Anomaly Detection Algorithms in Annotated Datasets from the Maritime Domain
- 1 January 2016
- Published by INSTICC
- p. 250-257
Anomaly Detection using B-spline Control Points as Feature Space in Annotated Trajectory Data from the Maritime Domain
The detection of anomalies and outliers is an important task for surveillance applications as it supports operators in their decision making process. One major challenge for the operators is to keep focus and not to be overwhelmed by the amount of information supplied by different sensor systems. Therefore, helping an operator to identify important details in the incoming data stream is one possibility to strengthen their situation awareness. In order to achieve this aim, the operator needs a detection system with high accuracy and low false alarm rates, because only then the system can be trusted. Thus, a fast and reliable detection system based on b-spline representation is introduced. Each trajectory is estimated by its cubic b-spline representation. The normal behavior is then learned by different machine learning algorithm like support vector machines and artificial neural networks, and evaluated by using an annotated real dataset from the maritime domain. The results are compared to other algorithms.
- 1 January 2016
- Published by INSTICC
- p. 250-257
Anomaly Detection using B-spline Control Points as Feature Space in Annotated Trajectory Data from the Maritime Domain
The detection of anomalies and outliers is an important task for surveillance applications as it supports operators in their decision making process. One major challenge for the operators is to keep focus and not to be overwhelmed by the amount of information supplied by different sensor systems. Therefore, helping an operator to identify important details in the incoming data stream is one possibility to strengthen their situation awareness. In order to achieve this aim, the operator needs a detection system with high accuracy and low false alarm rates, because only then the system can be trusted. Thus, a fast and reliable detection system based on b-spline representation is introduced. Each trajectory is estimated by its cubic b-spline representation. The normal behavior is then learned by different machine learning algorithm like support vector machines and artificial neural networks, and evaluated by using an annotated real dataset from the maritime domain. The results are compared to other algorithms.