Pub Date : 2002-07-08DOI: 10.1109/ICIF.2002.1021230
P. Ainsleigh, T. Luginbuhl
The probabilistic multi-hypothesis tracking (PMHT) algorithm is extended for application to classification. The PMHT model is reformulated as a bank of continuous-state hidden Markov models, allowing for supervised learning of the class-conditional probability density models, and for likelihood evaluation of multicomponent signals.
{"title":"Multicomponent signal classification using the PMHT algorithm","authors":"P. Ainsleigh, T. Luginbuhl","doi":"10.1109/ICIF.2002.1021230","DOIUrl":"https://doi.org/10.1109/ICIF.2002.1021230","url":null,"abstract":"The probabilistic multi-hypothesis tracking (PMHT) algorithm is extended for application to classification. The PMHT model is reformulated as a bank of continuous-state hidden Markov models, allowing for supervised learning of the class-conditional probability density models, and for likelihood evaluation of multicomponent signals.","PeriodicalId":399150,"journal":{"name":"Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121738525","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-08DOI: 10.1109/ICIF.2002.1021189
A. Gad, M. Farooq
Various multisensor data fusion architectures have been utilized to support the Maritime Surveillance (MS) in maritime tactical and strategic operations. The military tactical situation is mechanized through data fusion thus improving the quality of target tracking system. One of the major problems is that the surveillance area is generally large, hence making it difficult to arrive at a feasible data fusion architecture. The latter arises due to timing, accuracy, and different types of sensors and sensor platforms. In this paper, various data fusion architectures for MS are discussed. The proposed system interacts with the data fusion processes at different information levels. This architecture is employed to support the MS for a typical maritime tactical scenario. The proposed architecture has an acceptable performance.
{"title":"Data fusion architecture for Maritime Surveillance","authors":"A. Gad, M. Farooq","doi":"10.1109/ICIF.2002.1021189","DOIUrl":"https://doi.org/10.1109/ICIF.2002.1021189","url":null,"abstract":"Various multisensor data fusion architectures have been utilized to support the Maritime Surveillance (MS) in maritime tactical and strategic operations. The military tactical situation is mechanized through data fusion thus improving the quality of target tracking system. One of the major problems is that the surveillance area is generally large, hence making it difficult to arrive at a feasible data fusion architecture. The latter arises due to timing, accuracy, and different types of sensors and sensor platforms. In this paper, various data fusion architectures for MS are discussed. The proposed system interacts with the data fusion processes at different information levels. This architecture is employed to support the MS for a typical maritime tactical scenario. The proposed architecture has an acceptable performance.","PeriodicalId":399150,"journal":{"name":"Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)","volume":"306 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124359736","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-08DOI: 10.1109/ICIF.2002.1020900
E. Besada-Portas, J. A. López-Orozco, Jesus M. de la Cruz
A multisensor fusion system that is used for estimating the location of a robot and the state of the objects around it is presented. The whole fusion system has been implemented as a dynamic Bayesian network (DBN) with the purpose of having a homogenous and formalized way of capturing the dependencies that exist between robot location, the state of the environment, and all sensorial data. At this stage of research it consists of two independent DBNs, one for estimating robot location and another for building an occupancy probabilistic map of the environment, which are the basis of a unified fusion system. The dependencies of the variables and information in the two DBNs are captured by a unique DBN constructed by adding arcs (and nodes if necessary) between the two DBNs. The DBN implemented so far can be used in robots with different sets of sensors.
{"title":"Unified fusion system based on Bayesian networks for autonomous mobile robots","authors":"E. Besada-Portas, J. A. López-Orozco, Jesus M. de la Cruz","doi":"10.1109/ICIF.2002.1020900","DOIUrl":"https://doi.org/10.1109/ICIF.2002.1020900","url":null,"abstract":"A multisensor fusion system that is used for estimating the location of a robot and the state of the objects around it is presented. The whole fusion system has been implemented as a dynamic Bayesian network (DBN) with the purpose of having a homogenous and formalized way of capturing the dependencies that exist between robot location, the state of the environment, and all sensorial data. At this stage of research it consists of two independent DBNs, one for estimating robot location and another for building an occupancy probabilistic map of the environment, which are the basis of a unified fusion system. The dependencies of the variables and information in the two DBNs are captured by a unique DBN constructed by adding arcs (and nodes if necessary) between the two DBNs. The DBN implemented so far can be used in robots with different sets of sensors.","PeriodicalId":399150,"journal":{"name":"Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)","volume":"225 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122361235","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-08DOI: 10.1109/ICIF.2002.1021009
Wang Ming-Hui, Peng Ying-ning, You Zhi-sheng
The joint probabilistic data association (JPDA) filter has a very good tracking performance in dense targets and heavy clutter environments. However, the JPDA filter also has a huge computer load and tends to combine neighboring tracks. In this paper, an improved JPDA algorithm is presented. The main feature of our method is improving the performance of the JPDA algorithm by improving the performance of the tracking gate. The effectiveness of this method is assessed by mathematical analysis.
{"title":"Improved joint probabilistic data association algorithm","authors":"Wang Ming-Hui, Peng Ying-ning, You Zhi-sheng","doi":"10.1109/ICIF.2002.1021009","DOIUrl":"https://doi.org/10.1109/ICIF.2002.1021009","url":null,"abstract":"The joint probabilistic data association (JPDA) filter has a very good tracking performance in dense targets and heavy clutter environments. However, the JPDA filter also has a huge computer load and tends to combine neighboring tracks. In this paper, an improved JPDA algorithm is presented. The main feature of our method is improving the performance of the JPDA algorithm by improving the performance of the tracking gate. The effectiveness of this method is assessed by mathematical analysis.","PeriodicalId":399150,"journal":{"name":"Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129853333","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-08DOI: 10.1109/ICIF.2002.1020891
Dongguang Zuo, Chongzhao Han, Zheng Lin, Hongyan Zhu, Han Hong
This paper develops a tracking algorithm for maneuvering target based on fuzzy logic inference (FMMTA). In place of the model probability computed intricately in the IMM, filtering measurement innovations are tackled with the innovation covariance, and the results are used as the input to a fuzzy inference system to get the matched degrees for each filtering model in the model set designed. With the matched degrees, the estimation from each filtering is weighted to obtain the maneuvering target's overall estimation and its covariance. The performance of FMMTA is tested via Monte Carlo simulation, and the result expresses its validity and its promise.
{"title":"Fuzzy multiple model tracking algorithm for manoeuvring target","authors":"Dongguang Zuo, Chongzhao Han, Zheng Lin, Hongyan Zhu, Han Hong","doi":"10.1109/ICIF.2002.1020891","DOIUrl":"https://doi.org/10.1109/ICIF.2002.1020891","url":null,"abstract":"This paper develops a tracking algorithm for maneuvering target based on fuzzy logic inference (FMMTA). In place of the model probability computed intricately in the IMM, filtering measurement innovations are tackled with the innovation covariance, and the results are used as the input to a fuzzy inference system to get the matched degrees for each filtering model in the model set designed. With the matched degrees, the estimation from each filtering is weighted to obtain the maneuvering target's overall estimation and its covariance. The performance of FMMTA is tested via Monte Carlo simulation, and the result expresses its validity and its promise.","PeriodicalId":399150,"journal":{"name":"Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)","volume":"01 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129880387","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-08DOI: 10.1109/ICIF.2002.1020968
J. Masters
Structured Knowledge Source Integration, or SKSI, is an ongoing research and development project at Cycorp intended to enable the Cyc knowledge base to integrate (access, query, assimilate, and merge) a variety of external structured knowledge sources, such as databases, spreadsheets, XML or DAML tagged text, GIS datasets, and queryable Web pages. With SKSI, the Cyc knowledge base will be able to draw upon information obtained from multiple knowledge sources when answering complex queries, to assimilate (transform and store) the contents of the knowledge sources directly into the Cyc knowledge base, and to mediate between several semantically similar knowledge sources. These capabilities will extend the flexibility and power of the Cyc product to serve as the universal ontology and knowledge repository in any application requiring knowledge based reasoning. This article discusses some of the main technical issues of knowledge source integration, reviews some of the literature on the subject, describes some elements of the SKSI approach, illustrates two example Cyc queries that use two structured knowledge sources already mapped into Cyc, and proposes a Schema Modeling Toolkit of applications we are designing to leverage the core SKSI development.
{"title":"Structured Knowledge Source Integration and its applications to information fusion","authors":"J. Masters","doi":"10.1109/ICIF.2002.1020968","DOIUrl":"https://doi.org/10.1109/ICIF.2002.1020968","url":null,"abstract":"Structured Knowledge Source Integration, or SKSI, is an ongoing research and development project at Cycorp intended to enable the Cyc knowledge base to integrate (access, query, assimilate, and merge) a variety of external structured knowledge sources, such as databases, spreadsheets, XML or DAML tagged text, GIS datasets, and queryable Web pages. With SKSI, the Cyc knowledge base will be able to draw upon information obtained from multiple knowledge sources when answering complex queries, to assimilate (transform and store) the contents of the knowledge sources directly into the Cyc knowledge base, and to mediate between several semantically similar knowledge sources. These capabilities will extend the flexibility and power of the Cyc product to serve as the universal ontology and knowledge repository in any application requiring knowledge based reasoning. This article discusses some of the main technical issues of knowledge source integration, reviews some of the literature on the subject, describes some elements of the SKSI approach, illustrates two example Cyc queries that use two structured knowledge sources already mapped into Cyc, and proposes a Schema Modeling Toolkit of applications we are designing to leverage the core SKSI development.","PeriodicalId":399150,"journal":{"name":"Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130364308","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-08DOI: 10.1109/ICIF.2002.1021228
S. Davey, D. Gray, S. Colegrove
An important problem in multi-target tracking is track initiation and termination. The tracking algorithm aims to discriminate false detections caused by various sources of interference from valid detections caused by targets of interest. This is a problem of model order estimation. One approach to solving this problem with the Probabilistic Data Association Filter has been referred to as target visibility. This paper shows how the target visibility model can be incorporated into the Probabilistic Multi-Hypothesis Tracker to provide integrated initiation and termination.
{"title":"A Markov model for initiating tracks with the probabilistic multi-hypothesis tracker","authors":"S. Davey, D. Gray, S. Colegrove","doi":"10.1109/ICIF.2002.1021228","DOIUrl":"https://doi.org/10.1109/ICIF.2002.1021228","url":null,"abstract":"An important problem in multi-target tracking is track initiation and termination. The tracking algorithm aims to discriminate false detections caused by various sources of interference from valid detections caused by targets of interest. This is a problem of model order estimation. One approach to solving this problem with the Probabilistic Data Association Filter has been referred to as target visibility. This paper shows how the target visibility model can be incorporated into the Probabilistic Multi-Hypothesis Tracker to provide integrated initiation and termination.","PeriodicalId":399150,"journal":{"name":"Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)","volume":"210 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128816450","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-08DOI: 10.1109/ICIF.2002.1021193
W. P. Malcolm, A. Doucet, S. Zollo
In this article we consider tracking a single maneuvering target in scenarios where range information is not available, or is denied. This tracking problem is usually referred to as passive ranging, or bearings-only tracking. Tracking any single maneuvering target naturally admits a jump Markov system, in which a collection of candidate dynamical systems is proposed to model various classes of motion, each of which is assumed to be executed by the target according to a Markov law. Standard techniques to solve this problem use the so called interacting multiple model (IMM), or its variants. Recently sequential Monte Carlo (SMC) techniques have been applied to passive ranging problems, however, most of the scenarios reported in the literature consider nonmaneuvering targets. In this article we apply a new SMC technique to the passive ranging problem in a maneuvering target scenario. The algorithm we propose is compared to the so called auxiliary particle filter (APF). A simulation study is included.
{"title":"Sequential Monte Carlo tracking schemes for maneuvering targets with passive ranging","authors":"W. P. Malcolm, A. Doucet, S. Zollo","doi":"10.1109/ICIF.2002.1021193","DOIUrl":"https://doi.org/10.1109/ICIF.2002.1021193","url":null,"abstract":"In this article we consider tracking a single maneuvering target in scenarios where range information is not available, or is denied. This tracking problem is usually referred to as passive ranging, or bearings-only tracking. Tracking any single maneuvering target naturally admits a jump Markov system, in which a collection of candidate dynamical systems is proposed to model various classes of motion, each of which is assumed to be executed by the target according to a Markov law. Standard techniques to solve this problem use the so called interacting multiple model (IMM), or its variants. Recently sequential Monte Carlo (SMC) techniques have been applied to passive ranging problems, however, most of the scenarios reported in the literature consider nonmaneuvering targets. In this article we apply a new SMC technique to the passive ranging problem in a maneuvering target scenario. The algorithm we propose is compared to the so called auxiliary particle filter (APF). A simulation study is included.","PeriodicalId":399150,"journal":{"name":"Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126181811","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-08DOI: 10.1109/ICIF.2002.1021202
Tanzeem Chaodhury, James M. Rehg, V. Pavlovic, A. Pentland
Bayesian networks are an attractive modeling tool for human sensing, as they combine an intuitive graphical representation with efficient algorithms for inference and learning. Temporal fusion of multiple sensors can be efficiently formulated using dynamic Bayesian networks (DBNs) which allow the power of statistical inference and learning to be combined with contextual knowledge of the problem. Unfortunately, simple learning methods can cause such appealing models to fail when the data exhibits complex behavior We first demonstrate how boosted parameter learning could be used to improve the performance of Bayesian network classifiers for complex multimodal inference problems. As an example we apply the framework to the problem of audiovisual speaker detection in an interactive environment using "off-the-shelf" visual and audio sensors (face, skin, texture, mouth motion, and silence detectors). We then introduce a boosted structure learning algorithm. Given labeled data, our algorithm modifies both the network structure and parameters so as to improve classification accuracy. We compare its performance to both standard structure learning and boosted parameter learning. We present results for speaker detection and for datasets from the UCI repository.
{"title":"Boosted learning in dynamic Bayesian networks for multimodal detection","authors":"Tanzeem Chaodhury, James M. Rehg, V. Pavlovic, A. Pentland","doi":"10.1109/ICIF.2002.1021202","DOIUrl":"https://doi.org/10.1109/ICIF.2002.1021202","url":null,"abstract":"Bayesian networks are an attractive modeling tool for human sensing, as they combine an intuitive graphical representation with efficient algorithms for inference and learning. Temporal fusion of multiple sensors can be efficiently formulated using dynamic Bayesian networks (DBNs) which allow the power of statistical inference and learning to be combined with contextual knowledge of the problem. Unfortunately, simple learning methods can cause such appealing models to fail when the data exhibits complex behavior We first demonstrate how boosted parameter learning could be used to improve the performance of Bayesian network classifiers for complex multimodal inference problems. As an example we apply the framework to the problem of audiovisual speaker detection in an interactive environment using \"off-the-shelf\" visual and audio sensors (face, skin, texture, mouth motion, and silence detectors). We then introduce a boosted structure learning algorithm. Given labeled data, our algorithm modifies both the network structure and parameters so as to improve classification accuracy. We compare its performance to both standard structure learning and boosted parameter learning. We present results for speaker detection and for datasets from the UCI repository.","PeriodicalId":399150,"journal":{"name":"Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)","volume":"146 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127571588","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-08DOI: 10.1109/ICIF.2002.1021144
J. R. Hoffman, R. Mahler
The concept of miss distance-Euclidean, Mahalanobis, etc.-is a fundamental, far-reaching, and taken-for-granted element of the engineering theory and practice of single-sensor, single-target systems. One might expect that multisensor, multitarget information fusion theory and applications would already rest upon a similarly fundamental concept-namely, miss distance between multi-object systems (i.e., systems in which not only individual objects can vary, but their number as well). However, this has not been the case. Consequently, in this paper we introduce a comprehensive theory of distance metrics for multitarget (and, more generally, multi-object) systems. We show that this theory extends an optimal-assignment approach proposed by O. Drummond. We describe tractable computational approaches for computing such metrics, as well as some potentially far-reaching implications for applications such as sensor management.
{"title":"Multitarget miss distance and its applications","authors":"J. R. Hoffman, R. Mahler","doi":"10.1109/ICIF.2002.1021144","DOIUrl":"https://doi.org/10.1109/ICIF.2002.1021144","url":null,"abstract":"The concept of miss distance-Euclidean, Mahalanobis, etc.-is a fundamental, far-reaching, and taken-for-granted element of the engineering theory and practice of single-sensor, single-target systems. One might expect that multisensor, multitarget information fusion theory and applications would already rest upon a similarly fundamental concept-namely, miss distance between multi-object systems (i.e., systems in which not only individual objects can vary, but their number as well). However, this has not been the case. Consequently, in this paper we introduce a comprehensive theory of distance metrics for multitarget (and, more generally, multi-object) systems. We show that this theory extends an optimal-assignment approach proposed by O. Drummond. We describe tractable computational approaches for computing such metrics, as well as some potentially far-reaching implications for applications such as sensor management.","PeriodicalId":399150,"journal":{"name":"Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)","volume":"496 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129278782","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：