| 1. |
- Blasch, Erik, et al.
(författare)
-
Implication of Culture : User Roles in Information Fusion for Enhanced Situational Understanding
- 2009
-
Ingår i: Information Fusion, 2009. FUSION '09. 12th International Conference on. - : IEEE conference proceedings. - 9780982443804 ; , s. 1272-1279
-
Konferensbidrag (refereegranskat)abstract
- Information Fusion coordinates large-volume data processing machines to address user needs. Users expect a situational picture to extend their ability of sensing events, movements, and activities. Typically, data is collected and processed for object location (e.g. target identification) and movement (e.g. tracking); however, high-level reasoning or situational understanding depends on the spatial, cultural, and political effects. In this paper, we explore opportunities where information fusion can aid in the selection and processing of the data for enhanced tacit knowledge understanding by (1) display fusion for data presentation (e.g. cultural segmentation), (2) interactive fusion to allow the user to inject a priori knowledge (e..g. cultural values), and (3) associated metrics of predictive capabilities (e.g. cultural networks). In a simple scenario, cultural information impacts on situational understanding from deception is demonstrated using the Technology-Emotion-Culture-Knowledge (TECK) attributes of the Observer-Orient-Decide-Act (OODA) model.
|
|
| 2. |
- Brax, Christoffer, et al.
(författare)
-
An ensemble approach for increased anomaly detection performance in video surveillance data
- 2009
-
Ingår i: Proceedings of the 12th International Conference on Information Fusion (FUSION 2009), Seattle, Washington, USA, 6–9 July 2009. - : IEEE conference proceedings. - 9780982443804 ; , s. 694-701
-
Konferensbidrag (refereegranskat)abstract
- The increased societal need for surveillance and the decrease in cost of sensors have led to a number of new challenges. The problem is not to collect data but to use it effectively for decision support. Manual interpretation of huge amounts of data in real-time is not feasible; the operator of a surveillance system needs support to analyze and understand all incoming data. In this paper an approach to intelligent video surveillance is presented, with emphasis on finding behavioural anomalies. Two different anomaly detection methods are compared and combined. The results show that it is possible to best increase the total detection performance by combining two different anomaly detectors rather than employing them independently.
|
|
| 3. |
- Deegalla, Sampath, et al.
(författare)
-
Fusion of Dimensionality Reduction Methods : a Case Study in Microarray Classification
- 2009
-
Ingår i: Proceedings of the 12th International Conference on Information Fusion. - : ISIF. - 9780982443804 ; , s. 460-465
-
Konferensbidrag (refereegranskat)abstract
- Dimensionality reduction has been demonstrated to improve the performance of the k-nearest neighbor (kNN) classifier for high-dimensional data sets, such as microarrays. However, the effectiveness of different dimensionality reduction methods varies, and it has been shown that no single method constantly outperforms the others. In contrast to using a single method, two approaches to fusing the result of applying dimensionality reduction methods are investigated: feature fusion and classifier fusion. It is shown that by fusing the output of multiple dimensionality reduction techniques, either by fusing the reduced features or by fusing the output of the resulting classifiers, both higher accuracy and higher robustness towards the choice of number of dimensions is obtained.
|
|
| 4. |
- Fooladvandi, Farzad, et al.
(författare)
-
Signature-based activity detection based on Bayesian networks acquired from expert knowledge
- 2009
-
Ingår i: Proceedings of the 12<sup>th</sup> International Conference on Information Fusion (FUSION 2009). - : ISIF. - 9780982443804 ; , s. 436-443
-
Konferensbidrag (refereegranskat)abstract
- The maritime industry is experiencing one of its longest and fastest periods of growth. Hence, the global maritime surveillance capacity is in a great need of growth as well. The detection of vessel activity is an important objective of the civil security domain. Detecting vessel activity may become problematic if audit data is uncertain. This paper aims to investigate if Bayesian networks acquired from expert knowledge can detect activities with a signature-based detection approach. For this, a maritime pilot-boat scenario has been identified with a domain expert. Each of the scenario’s activities has been divided up into signatures where each signature relates to a specific Bayesian network information node. The signatures were implemented to find evidences for the Bayesian network information nodes. AIS-data with real world observations have been used for testing, which have shown that it is possible to detect the maritime pilot-boat scenario based on the taken approach.
|
|
| 5. |
- Hilletofth, Per, et al.
(författare)
-
Agent-Based Simulation Fusion for Improved Decision Making for Service Operations
- 2009
-
Ingår i: Proceedings of the 12th International Conference on Information Fusion. - : IEEE. - 9780982443804
-
Konferensbidrag (refereegranskat)abstract
- We use agent-based modeling and simulation to fuse data from multiple sources to estimate the state of some system properties. This implies that the real system of interest is modeled and simulated using agent principles. Using Monte-Carlo simulation, we estimate the values of some decision-relevant numerical properties, such as utilization of resources and service levels, as a decision support for a Maintenance Service Provider. Our initial results indicate that this kind of fusion of information sources can improve the understanding of the problem domain (e.g. to what degree some critical properties influence service operations) and also generate a basis for decision-making.
|
|
| 6. |
- Hilletofth, Per, et al.
(författare)
-
Agent-based simulation fusion for improved decision making for service operations
- 2009
-
Ingår i: Proceedings of the 12th International Conference on Information Fusion. - Seattle, US. - 9780982443804 ; , s. 998-1005
-
Konferensbidrag (refereegranskat)abstract
- We use agent-based modeling and simulation to fuse data from multiple sources to estimate the state of some system properties. This implies that the real system of interest is modeled and simulated using agent principles. Using Monte-Carlo simulation, we estimate the values of some decision-relevant numerical properties. We use the estimated properties, such as utilization of resources and service levels, as a decision support for a Maintenance Service Provider. Our initial results indicate that this kind of fusion of information sources can improve the understanding of the problem domain (e.g. to what degree some critical properties influence service operations) and also generate a basis for decision-making.
|
|
| 7. |
- Laxhammar, Rikard, et al.
(författare)
-
Anomaly detection in sea traffic - a comparison of the Gaussian Mixture Model and the Kernel Density Estimator
- 2009
-
Ingår i: Proceedings of the 12th International Conference on Information Fusion. - : ISIF. - 9780982443804 ; , s. 756-763
-
Konferensbidrag (refereegranskat)abstract
- This paper presents a first attempt to evaluate two previously proposed methods for statistical anomaly detection in sea traffic, namely the Gaussian Mixture Model (GMM) and the adaptive Kernel Density Estimator (KDE). A novel performance measure related to anomaly detection, together with an intermediate performance measure related to normalcy modeling, are proposed and evaluated using recorded AIS data of vessel traffic andsimulated anomalous trajectories. The normalcy modeling evaluation indicates that KDE more accurately captures finer details of normal data. Yet, results from anomaly detection show no significant difference between the two techniques and the performance of both is considered suboptimal. Part of the explanation is that the methods are based on a rather artificial division of data into geographical cells. The paper therefore discusses other clustering approaches based on more informed features of data and more background knowledge regarding the structure and natural classes of the data.
|
|
| 8. |
- Lindgren, David, et al.
(författare)
-
Shooter Localization in Wireless Sensor Networks
- 2009
-
Ingår i: Proceedings of the 12th International Conference on Information Fusion. - 9780982443804 ; , s. 404-411
-
Konferensbidrag (refereegranskat)abstract
- Shooter localization is considered in a wireless network of microphones. Both the acoustic muzzle blast (MB) from the gunfire and the ballistic shock wave (SW) from the bullet can be detected by the microphones and are considered as measurements. The MB measurements give rise to a standard sensor net-work problem, similar to time difference of arrivals in cellular phone networks, and the localization accuracy is good provided that the sensors are well synchronized compared to the MB detection accuracy. The detection times of the SW depend on both shooter position and aiming angle, and we demonstrate that estimation based on these measurements can potentially also reveal the shooting direction beside the position, but again this re-quires good synchronization. We propose to base the estimation on the time difference of MB and SW at each sensor, which becomes insensitive to synchroniza-tion. Cramér-Rao lower bound analysis indicates how a lower bound of the root mean square error depends on the synchronization error for MB and MB-SW differ-ences, respectively. Results from field trials with differ-ent type of ammunition show excellent accuracy of the proposed method for both the position and the aiming angle of the shooter.
|
|
| 9. |
- Orguner, Umut, et al.
(författare)
-
Distributed Target Tracking with Propagation Delayed Measurements
- 2009
-
Ingår i: Proceedings of the 12th International Conference on Information Fusion. - 9780982443804 ; , s. 802-809
-
Konferensbidrag (refereegranskat)abstract
- This paper presents a framework for making distributed target tracking under significant signal propagation delays between the target and the sensors. Each sensor considered makes estimation using its own measurements compensating for the involved signal propagation delay using a deterministic sampling based algorithm proposed previously. Since the individual sensor readings might not be enough to localize the target, the sensors have to share their estimates with each other at specific time instants and correct their individual estimates. This work is mainly related to how this estimate correction and fusion should be carried out. An internal covariance approximation which keeps consistency but at the same time bypasses the track correlation problem is proposed. The results are illustrated on a challenging two-sensor bearings-only tracking scenario.
|
|
| 10. |
- Savic, Vladimir, et al.
(författare)
-
Sensor localization using nonparametric generalized belief propagation in network with loop
- 2009
-
Ingår i: IEEE Proc. of Intl. Conf. on Information Fusion (FUSION). - 9780982443804
-
Konferensbidrag (refereegranskat)abstract
- Belief propagation (BP) is one of the best-known graphical model for inference in statistical physics, artificial intelligence, computer vision, etc. Furthermore, a recent research in distributed sensor network localization showed us that BP is an efficient way to obtain sensor location as well as appropriate uncertainty. However, BP convergence is not guaranteed in a network with loops. In this paper, we propose localization using generalized belief propagation based on junction tree method (GBP-JT) and nonparametric (particle-based) approximation of this algorithm (NGBP-JT). We illustrate it in a network with loop where BP shows poor performance. In fact, we compared estimated locations with nonparametric belief propagation (NBP) algorithm. According to our simulation results, GBP-JT resolved the problems with loops, but the price for this is unacceptable large computational cost. Therefore, our approximated version of this algorithm, NGBP-JT, reduced significantly this cost, with little effect on accuracy.
|
|
| 11. |
- Skoglar, Per, et al.
(författare)
-
Road Target Tracking with an Approximative Rao-Blackwellized Particle Filter
- 2009
-
Ingår i: Proceedings from the 12th International Conference on Information Fusion, 6-9 July, Seattle, Washington, USA. - : IEEE conference proceedings. - 9780982443804 ; , s. 17-24
-
Konferensbidrag (refereegranskat)abstract
- Using prior information about the road network will improve the estimation performance for a road constrained target significantly. Several estimation methods have been proposed to handle the multi-modality that arise in a road target tracking application. One popular filter suitable for this kind of non-linear problems is the Particle Filter, but a major drawback is that the Particle filter requires a large amount of particles as the state dimension increases to maintain a good approximation of the filtering distribution. In this paper a Rao-Blackwellized Particle Filter based approach is proposed to reduce the dimension of the state space in road target tracking applications. Furthermore, it is also shown how prior information about the probability of detection can be used to improve the estimation performance further.
|
|
| 12. |
- Svensson, Daniel, 1979, et al.
(författare)
-
Performance Evaluation of MHT and GM-CPHD in a Ground Target Tracking Scenario
- 2009
-
Ingår i: Proceedings of the 12th International Conference on Information Fusion. - 9780982443804 ; , s. 300-307
-
Konferensbidrag (refereegranskat)abstract
- Performance evaluations of multi-target tracking algorithms are often limited to consider comparisons within the same algorithm family. In this paper, two conceptually different multi-target tracking algorithms are evaluated, namely a multiple-hypothesis tracking (MHT) algorithm and the Gaussian mixture cardinalized probability hypothesis density (GM-CPHD) filter. As a reference, a conventional single-hypothesis tracking algorithm is included in the evaluation. The performance is assessed using the root-mean square error of the estimated number of targets, and the recently published optimal subpattern assignment (OSPA) measure. The scenario under consideration is tracking of nine closely spaced ground targets, using simulated measurements from an airborne radar. By observing the estimation of the number of targets, as well as of the target states, conclusions are drawn regarding the behavior of MHT and GM-CPHD. For example, GM-CPHD is more responsive to changes in the number of targets, whereas MHT is less responsive, but produces a more stable output.
|
|
| 13. |
- Svensson, Lennart, 1976, et al.
(författare)
-
Set JPDA algorithm for tracking unordered sets of targets
- 2009
-
Ingår i: Proceedings of the 12th International Conference on Information Fusion. - 9780982443804 ; , s. 1187-1194
-
Konferensbidrag (refereegranskat)abstract
- In this article we show that traditional tracking algorithms should be adjusted when the objective is to recursively estimate an unordered (unlabeled) set of target state vectors, i.e., when it is not of importance to try to preserve target identities over time. We study scenarios where the number of targets is known, and propose a new version of the joint probabilistic data association (JPDA) filter that we call set JPDA (SJPDA). Simulations show that the new filter outperforms the JPDA in a two-target scenario when evaluated according to the mean optimal subpattern assignment (MOSPA) measure.
|
|
| 14. |
- van Laere, Joeri
(författare)
-
Challenges for IF performance evaluation in practice
- 2009
-
Ingår i: Information Fusion, 2009. FUSION '09. 12th International Conference on. - : IEEE conference proceedings. - 9780982443804 ; , s. 866-873
-
Konferensbidrag (refereegranskat)abstract
- This study presents a literature analysis of 52 publications on the topic of information fusion performance evaluation. The aim is to identify whether information fusion researchers are aware of the problematic nature of performance evaluation in practice, which is substantially different from performance evaluation in test environments. As expected, only 6% of the publications do actually report on studies conducted in practice, and therefore more such studies are needed. However, 58% of the publications do critically reflect on the problematic nature of performance evaluation in practice. As ground truth is lacking in practice, there is a need for (a set of) performance measures which are both comprehensive and adaptable to domain or situation context and changing circumstances over time. More studies need to address the overall question how multiple performance measures can function in concert and how exactly can be judged whether adaptation has been an improvement or not.
|
|
| 15. |
- van Laere, Joeri, et al.
(författare)
-
Evaluation of a workshop to capture knowledge from subject matter experts in maritime surveillance
- 2009
-
Ingår i: Information Fusion, 2009. FUSION '09. 12th International Conference on. - : IEEE conference proceedings. - 9780982443804 ; , s. 171-178
-
Konferensbidrag (refereegranskat)abstract
- This paper reports on a brainstorming workshop with subject matter experts that aimed at identifying potential anomalies for world wide maritime surveillance. An extensive description of the workshop process and output is presented. The evaluation consists of two parts. First, the output of the workshop is compared with the output of a similar workshop as presented in Roy [1]. Secondly, the methodological differences between capturing know-how of subject matter experts in a workshop versus observing daily work of operators, as applied in Nilsson et al [2], is discussed. The conclusions are that brainstorming workshops and field studies should be combined to compensate for their weaknesses. Also, the outcomes of both methods heavily depend on the selection of operators to be observed, respectively the selection of subject matter experts invited. A final recommendation is that multiple studies and workshops are needed until no ‘new’ anomalies are identified.
|
|
