Pub Date : 2007-12-26DOI: 10.1109/ICIF.2007.4408000
T. Tatschke, Franziska Färber, E. Fuchs, Leonhard F. Walchshäusl, R. Lindl
In the development phase of perception systems (e.g. for advanced driver assistance systems) general interest is pointing towards the performance of the respective detection and tracking algorithms. One common way to evaluate such systems relies on simulated data which is used as a reference. We present a semi-autonomous method, which allows the extraction of reference data from sensor recordings (including data at least from a camera and a distance measuring sensor device). Furthermore, we show how to combine these reference data with the output from the object detection system and how to derive performance statistics (detection and miss rates) of the system. As the generated reference information can be stored along with the sensor recordings, this method also facilitates the comparison of different software versions or algorithm parameters.
{"title":"Semi-autonomous reference data generation for perception performance evaluation","authors":"T. Tatschke, Franziska Färber, E. Fuchs, Leonhard F. Walchshäusl, R. Lindl","doi":"10.1109/ICIF.2007.4408000","DOIUrl":"https://doi.org/10.1109/ICIF.2007.4408000","url":null,"abstract":"In the development phase of perception systems (e.g. for advanced driver assistance systems) general interest is pointing towards the performance of the respective detection and tracking algorithms. One common way to evaluate such systems relies on simulated data which is used as a reference. We present a semi-autonomous method, which allows the extraction of reference data from sensor recordings (including data at least from a camera and a distance measuring sensor device). Furthermore, we show how to combine these reference data with the output from the object detection system and how to derive performance statistics (detection and miss rates) of the system. As the generated reference information can be stored along with the sensor recordings, this method also facilitates the comparison of different software versions or algorithm parameters.","PeriodicalId":298941,"journal":{"name":"2007 10th International Conference on Information Fusion","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-12-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116687321","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 : 2007-12-01DOI: 10.1109/CDC.2007.4434638
Huimin Chen, Y. Bar-Shalom
Summary form only given. The problem of track-to-track association and track fusion has been considered in the literature where the local trackers assume the same target motion model and send their local state estimates to the fusion center on demand. Many issues arise when local trackers use different target motion models or even operate on different target state spaces. In this case, selecting an appropriate track association method at the fusion center is essential to the performance of the overall tracking system. In this paper, we examine several track association methods with different assumptions on the target distribution in a surveillance region. We found that the track association performance can be very different among these methods even when they have the same desired significance level of the correct association probability. We recommend to use the track association method that best approximates the likelihood ratio test with complete knowledge of the target distribution. In addition, existing track fusion techniques have to be modified to account for the model mismatch among some of the local trackers. The track association and fusion problem is illustrated by a two dimensional tracking example with both radar tracks and electronic support measures (ESM) tracks.
{"title":"Track association and fusion with heterogeneous local trackers","authors":"Huimin Chen, Y. Bar-Shalom","doi":"10.1109/CDC.2007.4434638","DOIUrl":"https://doi.org/10.1109/CDC.2007.4434638","url":null,"abstract":"Summary form only given. The problem of track-to-track association and track fusion has been considered in the literature where the local trackers assume the same target motion model and send their local state estimates to the fusion center on demand. Many issues arise when local trackers use different target motion models or even operate on different target state spaces. In this case, selecting an appropriate track association method at the fusion center is essential to the performance of the overall tracking system. In this paper, we examine several track association methods with different assumptions on the target distribution in a surveillance region. We found that the track association performance can be very different among these methods even when they have the same desired significance level of the correct association probability. We recommend to use the track association method that best approximates the likelihood ratio test with complete knowledge of the target distribution. In addition, existing track fusion techniques have to be modified to account for the model mismatch among some of the local trackers. The track association and fusion problem is illustrated by a two dimensional tracking example with both radar tracks and electronic support measures (ESM) tracks.","PeriodicalId":298941,"journal":{"name":"2007 10th International Conference on Information Fusion","volume":"32 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116126206","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 : 2007-07-09DOI: 10.1109/ICIF.2007.4408213
B. Rhodes
Contemporary situational awareness problems such as automated normalcy learning for anomaly detection and motion behavior prediction are addressed with biologically-inspired processing, representation, and learning approaches. Issues and challenges are discussed and our responses to them described. Relatively simple neural principles provide considerable power in providing capabilities required to learn models of normal motion behavior and utilize those models to identify unusual behavior or determine the most likely future behavior of objects of interest.
{"title":"Biologically-inspired approaches to higher-level information fusion","authors":"B. Rhodes","doi":"10.1109/ICIF.2007.4408213","DOIUrl":"https://doi.org/10.1109/ICIF.2007.4408213","url":null,"abstract":"Contemporary situational awareness problems such as automated normalcy learning for anomaly detection and motion behavior prediction are addressed with biologically-inspired processing, representation, and learning approaches. Issues and challenges are discussed and our responses to them described. Relatively simple neural principles provide considerable power in providing capabilities required to learn models of normal motion behavior and utilize those models to identify unusual behavior or determine the most likely future behavior of objects of interest.","PeriodicalId":298941,"journal":{"name":"2007 10th International Conference on Information Fusion","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124290972","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 : 2007-07-09DOI: 10.1109/ICIF.2007.4407990
Panzhi Liu, Chongzhao Han, Ming Lei, Zengguo Sun
This paper present a new distributed CFAR (constant false alarm rate) detector based on the adaptive censored cell-averaging CFAR technique. In the scheme, every local decision of individual detector, resulting from the comparison between its sample in test cell and the estimate of clutter power level of its reference samples, takes the value zero or one. In local processor, the CCA-CFAR (censored cell-averaging) technique is utilized to get the local decision, Then, the fusion center makes the global decision based on the total local decisions, which are transmitted from each local sensor. The overall decision, which is zero or one, is obtained at the data fusion center grounded on "k/N" fusion rule. The results show that for the nonhomogeneous background caused by multiple interfering targets, this approach is more reality. Particularly in multiple target situations, it exhibits robustness than MOS (maximum order statistic), mOS (minimum order statistic), and OSOR (ordered statistics), ORAND in distributed sensor networks. Under Swerling 2 assumption, the analytic expression of false alarm and detection probability are derived.
{"title":"Adaptive censored cell-averaging CFAR detection in distributed sensor networks","authors":"Panzhi Liu, Chongzhao Han, Ming Lei, Zengguo Sun","doi":"10.1109/ICIF.2007.4407990","DOIUrl":"https://doi.org/10.1109/ICIF.2007.4407990","url":null,"abstract":"This paper present a new distributed CFAR (constant false alarm rate) detector based on the adaptive censored cell-averaging CFAR technique. In the scheme, every local decision of individual detector, resulting from the comparison between its sample in test cell and the estimate of clutter power level of its reference samples, takes the value zero or one. In local processor, the CCA-CFAR (censored cell-averaging) technique is utilized to get the local decision, Then, the fusion center makes the global decision based on the total local decisions, which are transmitted from each local sensor. The overall decision, which is zero or one, is obtained at the data fusion center grounded on \"k/N\" fusion rule. The results show that for the nonhomogeneous background caused by multiple interfering targets, this approach is more reality. Particularly in multiple target situations, it exhibits robustness than MOS (maximum order statistic), mOS (minimum order statistic), and OSOR (ordered statistics), ORAND in distributed sensor networks. Under Swerling 2 assumption, the analytic expression of false alarm and detection probability are derived.","PeriodicalId":298941,"journal":{"name":"2007 10th International Conference on Information Fusion","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117348657","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 : 2007-07-09DOI: 10.1109/ICIF.2007.4408090
Akrem El-ghazal, O. Basir, S. Belkasim
Shape-based image retrieval techniques are among the most successful content-based image retrieval (CBIR) approaches. In recent years, the number of shape-based image retrieval techniques has dramatically increased; however, each technique has both advantages and shortcomings. This paper proposes a consensus-based fusion algorithm to integrate several shape-based image retrieval techniques so as to enhance the performance of the image retrieval process. In this algorithm, several techniques work as a team: they exchange their ranking information based on pair-wise co-ranking to reach a consensus that will improve their final ranking decisions. Although the proposed algorithm handles any number of CBIR techniques, only three common techniques are used to demonstrate its effectiveness. Several experiments were conducted on the widely used MPEG-7 database. The results indicate that the proposed fusion algorithm significantly improves the retrieval process.
{"title":"A consensus-based fusion algorithm in shape-based image retrieval","authors":"Akrem El-ghazal, O. Basir, S. Belkasim","doi":"10.1109/ICIF.2007.4408090","DOIUrl":"https://doi.org/10.1109/ICIF.2007.4408090","url":null,"abstract":"Shape-based image retrieval techniques are among the most successful content-based image retrieval (CBIR) approaches. In recent years, the number of shape-based image retrieval techniques has dramatically increased; however, each technique has both advantages and shortcomings. This paper proposes a consensus-based fusion algorithm to integrate several shape-based image retrieval techniques so as to enhance the performance of the image retrieval process. In this algorithm, several techniques work as a team: they exchange their ranking information based on pair-wise co-ranking to reach a consensus that will improve their final ranking decisions. Although the proposed algorithm handles any number of CBIR techniques, only three common techniques are used to demonstrate its effectiveness. Several experiments were conducted on the widely used MPEG-7 database. The results indicate that the proposed fusion algorithm significantly improves the retrieval process.","PeriodicalId":298941,"journal":{"name":"2007 10th International Conference on Information Fusion","volume":"45 22","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120965828","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 : 2007-07-09DOI: 10.1109/ICIF.2007.4408186
C. Berger, M. Daun, W. Koch
Target tracking from incomplete measurements of distinct sensors in a sensor network is a task of data fusion, present in a lot of applications. Difficulties in tracking using extended Kalman filters lead to unstable behavior, mainly caused by difficult initialization. Instead of using numerical batch-estimators, we offer an analytical approach to initialize the filter from a minimum number of observations. Additionally, we provide the possibility to estimate only sub-sets of parameters, and to reliably model resulting added uncertainties by the covariance matrix. The approach will be studied in two practical examples: 3D track initialization using bearings-only measurements and using slant-range and azimuth only. Numerical results will include performance and consistency analysis via Monte-Carlo simulations and comparison to the Cramer-Rao lower bound.
{"title":"Track initialization from incomplete measurements","authors":"C. Berger, M. Daun, W. Koch","doi":"10.1109/ICIF.2007.4408186","DOIUrl":"https://doi.org/10.1109/ICIF.2007.4408186","url":null,"abstract":"Target tracking from incomplete measurements of distinct sensors in a sensor network is a task of data fusion, present in a lot of applications. Difficulties in tracking using extended Kalman filters lead to unstable behavior, mainly caused by difficult initialization. Instead of using numerical batch-estimators, we offer an analytical approach to initialize the filter from a minimum number of observations. Additionally, we provide the possibility to estimate only sub-sets of parameters, and to reliably model resulting added uncertainties by the covariance matrix. The approach will be studied in two practical examples: 3D track initialization using bearings-only measurements and using slant-range and azimuth only. Numerical results will include performance and consistency analysis via Monte-Carlo simulations and comparison to the Cramer-Rao lower bound.","PeriodicalId":298941,"journal":{"name":"2007 10th International Conference on Information Fusion","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121277112","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 : 2007-07-09DOI: 10.1109/ICIF.2007.4407972
P. Abeles, M. Kovacich
Tracking maneuvering targets is a difficult problem due to unpredictable maneuvers which change the target's state and/or dynamics. To ensure track accuracy a filter needs to model the target correctly and quickly respond to maneuvers. A new sequential filter is proposed which attempts to improve upon existing algorithms in several areas. A more flexible internal model is used to describe effects of maneuver events. Maneuver hypotheses have improved temporal accuracy. A semi-Markov process is used to describe the probability of an event occurring as a function of time. In simulated test scenarios the new algorithm performs as well as or significantly better than Interacting Multiple Model filter.
{"title":"A semi-Markov multiple event filter for maneuvering targets","authors":"P. Abeles, M. Kovacich","doi":"10.1109/ICIF.2007.4407972","DOIUrl":"https://doi.org/10.1109/ICIF.2007.4407972","url":null,"abstract":"Tracking maneuvering targets is a difficult problem due to unpredictable maneuvers which change the target's state and/or dynamics. To ensure track accuracy a filter needs to model the target correctly and quickly respond to maneuvers. A new sequential filter is proposed which attempts to improve upon existing algorithms in several areas. A more flexible internal model is used to describe effects of maneuver events. Maneuver hypotheses have improved temporal accuracy. A semi-Markov process is used to describe the probability of an event occurring as a function of time. In simulated test scenarios the new algorithm performs as well as or significantly better than Interacting Multiple Model filter.","PeriodicalId":298941,"journal":{"name":"2007 10th International Conference on Information Fusion","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127445510","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 : 2007-07-09DOI: 10.1109/ICIF.2007.4408044
K. M. Fauske, F. Gustafsson, Øyvind Hegrenæs
This paper presents a method for identifying dynamic models of autonomous underwater vehicles (AUV) from logged data and a physically motivated model structure. Such models are instrumental for model-based control system design, but also for integrated navigation systems. We motive our work from the perspective of developing second generation integrated navigation systems, which use a sensor fusion approach to merge external information with a dynamic model for purposes of redundancy, integrity, and for fault detection and isolation.
{"title":"Estimation of AUV dynamics for sensor fusion","authors":"K. M. Fauske, F. Gustafsson, Øyvind Hegrenæs","doi":"10.1109/ICIF.2007.4408044","DOIUrl":"https://doi.org/10.1109/ICIF.2007.4408044","url":null,"abstract":"This paper presents a method for identifying dynamic models of autonomous underwater vehicles (AUV) from logged data and a physically motivated model structure. Such models are instrumental for model-based control system design, but also for integrated navigation systems. We motive our work from the perspective of developing second generation integrated navigation systems, which use a sensor fusion approach to merge external information with a dynamic model for purposes of redundancy, integrity, and for fault detection and isolation.","PeriodicalId":298941,"journal":{"name":"2007 10th International Conference on Information Fusion","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125119537","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 : 2007-07-09DOI: 10.1109/ICIF.2007.4408080
Xuezhi Wang, D. Musicki
Energy based detection measures sensor received signal strength (RSS) transmitted from a target. In this paper, we propose a new approach for estimating a moving target trajectory over a sensor field via energy based detections as an alternative to trilateration positioning or nonlinear estimation. In 2D case, possible target locations described by a RSS ratio from two sensors are approximated using a set of Gaussian random variables which are refereed to as location measurements. At each data collection time, several sets of such measurements can be found from RSS ratios which are due to multiple sensor detections. A track splitting filter is used to perform either measurement fusion and target state estimation using these measurements. The RSS ratio data mapping via Gaussian density approximation plays a key role in the proposed target tracking method and is robust in the sense that it can tolerate larger RSS noise and using additional sensor detections to improve tracking performance over trilateration based techniques. The effectiveness of the propose method is demonstrated via an example of tracking a moving target over a sensor network of small acoustic sensors.
{"title":"Target tracking using energy based detections","authors":"Xuezhi Wang, D. Musicki","doi":"10.1109/ICIF.2007.4408080","DOIUrl":"https://doi.org/10.1109/ICIF.2007.4408080","url":null,"abstract":"Energy based detection measures sensor received signal strength (RSS) transmitted from a target. In this paper, we propose a new approach for estimating a moving target trajectory over a sensor field via energy based detections as an alternative to trilateration positioning or nonlinear estimation. In 2D case, possible target locations described by a RSS ratio from two sensors are approximated using a set of Gaussian random variables which are refereed to as location measurements. At each data collection time, several sets of such measurements can be found from RSS ratios which are due to multiple sensor detections. A track splitting filter is used to perform either measurement fusion and target state estimation using these measurements. The RSS ratio data mapping via Gaussian density approximation plays a key role in the proposed target tracking method and is robust in the sense that it can tolerate larger RSS noise and using additional sensor detections to improve tracking performance over trilateration based techniques. The effectiveness of the propose method is demonstrated via an example of tracking a moving target over a sensor network of small acoustic sensors.","PeriodicalId":298941,"journal":{"name":"2007 10th International Conference on Information Fusion","volume":"94 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123244455","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 : 2007-07-09DOI: 10.1109/ICIF.2007.4408030
Y. Villers
This paper presents a fusion process developed for the future armoured vehicle system (FAVS) technical demonstration (TD) project. One of the project objectives was to develop, optimize and demonstrate a multi-sensor suite mounted on an army vehicle to detect and identify targets while the platform was moving. The sensors consisted of a cooled infrared camera, millimetre-wave radar and a defensive aide suite. For the purpose of this paper, only the first two sensors were considered in the fusion study. This paper presents the sensor suite, the fusion process, a description of a field trial used to gather target detection data, and the results and analysis of the fusion process.
{"title":"A fusion study of the FAVS sensors suite","authors":"Y. Villers","doi":"10.1109/ICIF.2007.4408030","DOIUrl":"https://doi.org/10.1109/ICIF.2007.4408030","url":null,"abstract":"This paper presents a fusion process developed for the future armoured vehicle system (FAVS) technical demonstration (TD) project. One of the project objectives was to develop, optimize and demonstrate a multi-sensor suite mounted on an army vehicle to detect and identify targets while the platform was moving. The sensors consisted of a cooled infrared camera, millimetre-wave radar and a defensive aide suite. For the purpose of this paper, only the first two sensors were considered in the fusion study. This paper presents the sensor suite, the fusion process, a description of a field trial used to gather target detection data, and the results and analysis of the fusion process.","PeriodicalId":298941,"journal":{"name":"2007 10th International Conference on Information Fusion","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116039711","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：