Pub Date : 2007-07-09DOI: 10.1109/ICIF.2007.4408011
Yuxi Chen, Chongzhao Han
We propose a hybrid method of seeded region growing and region hue-area information fusion for object segmentation under patterned background. At first, image is segmented into many small regions according to hue homogeneity by seeded region growing algorithm, then background texture mode is discovered by the regions' hue-area information fusion, finally, the background texture is removed according to the discovered background mode and seeded region growing, the residual region is the object segmentation result. Experimental results show that the method can be used in object segmentation under patterned background.
{"title":"A hybrid method of seeded region growing and region hue-area information fusion for object segmentation under patterned background","authors":"Yuxi Chen, Chongzhao Han","doi":"10.1109/ICIF.2007.4408011","DOIUrl":"https://doi.org/10.1109/ICIF.2007.4408011","url":null,"abstract":"We propose a hybrid method of seeded region growing and region hue-area information fusion for object segmentation under patterned background. At first, image is segmented into many small regions according to hue homogeneity by seeded region growing algorithm, then background texture mode is discovered by the regions' hue-area information fusion, finally, the background texture is removed according to the discovered background mode and seeded region growing, the residual region is the object segmentation result. Experimental results show that the method can be used in object segmentation under patterned background.","PeriodicalId":298941,"journal":{"name":"2007 10th International Conference on Information Fusion","volume":"53 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":"116380718","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.2007.4408004
Y. Boers, H. Driessen
Recently several new results for Cramer-Rao lower bounds (CRLB's) in dynamical systems have been obtained. Several different approaches and approximations have been presented. For the general case of target tracking with a detection probability smaller than one and possibly in the presence of false measurements, two main approaches have been presented. One is the so called information reduction factor (IRF) approach and the other the enumeration (ENUM) approach, also referred to as conditioning approach. It has been shown that the ENUM approach leads to a strictly larger covariance matrix than the IRF approach, still being a lower bound of on the performance however. Thus, the ENUM approach provides a strictly tighter bound on the attainable performance. It has been conjectured that these bounds converge to one another in the limit or equivalently after an initial transition stage. In this paper we show, using some recent results on the so called modified Riccati (MR) equation and by means of counter examples, that this conjecture does not hold true in general. We also prove that it does hold true in the special case of deterministic target motion. Furthermore, we show that the detection probability has an influence on the limiting behaviors of the bounds. The various results are illustrated by means of representative examples.
{"title":"Bounds for target tracking accuracy with probability of detection smaller than one","authors":"Y. Boers, H. Driessen","doi":"10.1109/ICIF.2007.4408004","DOIUrl":"https://doi.org/10.1109/ICIF.2007.4408004","url":null,"abstract":"Recently several new results for Cramer-Rao lower bounds (CRLB's) in dynamical systems have been obtained. Several different approaches and approximations have been presented. For the general case of target tracking with a detection probability smaller than one and possibly in the presence of false measurements, two main approaches have been presented. One is the so called information reduction factor (IRF) approach and the other the enumeration (ENUM) approach, also referred to as conditioning approach. It has been shown that the ENUM approach leads to a strictly larger covariance matrix than the IRF approach, still being a lower bound of on the performance however. Thus, the ENUM approach provides a strictly tighter bound on the attainable performance. It has been conjectured that these bounds converge to one another in the limit or equivalently after an initial transition stage. In this paper we show, using some recent results on the so called modified Riccati (MR) equation and by means of counter examples, that this conjecture does not hold true in general. We also prove that it does hold true in the special case of deterministic target motion. Furthermore, we show that the detection probability has an influence on the limiting behaviors of the bounds. The various results are illustrated by means of representative examples.","PeriodicalId":298941,"journal":{"name":"2007 10th International Conference on Information Fusion","volume":"53 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":"122668477","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.2007.4408100
A. Hanselmann, O. C. Schrempf, U. Hanebeck
In this paper, we present a novel approach to parametric density estimation from given samples. The samples are treated as a parametric density function by means of a Dirac mixture, which allows for applying analytic optimization techniques. The method is based on minimizing a distance measure between the integral of the approximation function and the empirical cumulative distribution function (EDF) of the given samples, where the EDF is represented by the integral of the Dirac mixture. Since this minimization problem cannot be solved directly in general, a progression technique is applied. Increased performance of the approach in comparison to iterative maximum likelihood approaches is shown in simulations.
{"title":"Optimal parametric density estimation by minimizing an analytic distance measure","authors":"A. Hanselmann, O. C. Schrempf, U. Hanebeck","doi":"10.1109/ICIF.2007.4408100","DOIUrl":"https://doi.org/10.1109/ICIF.2007.4408100","url":null,"abstract":"In this paper, we present a novel approach to parametric density estimation from given samples. The samples are treated as a parametric density function by means of a Dirac mixture, which allows for applying analytic optimization techniques. The method is based on minimizing a distance measure between the integral of the approximation function and the empirical cumulative distribution function (EDF) of the given samples, where the EDF is represented by the integral of the Dirac mixture. Since this minimization problem cannot be solved directly in general, a progression technique is applied. Increased performance of the approach in comparison to iterative maximum likelihood approaches is shown in simulations.","PeriodicalId":298941,"journal":{"name":"2007 10th International Conference on Information Fusion","volume":"100 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":"122683850","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.2007.4408136
S. Benferhat, S. Lagrue, J. Rossit
Recently, several approaches have been proposed to merge possibly contradictory belief bases. This paper focuses on max-based merging operators applied to incommensurable ranked belief bases. We first propose a characterization of a result of merging using Pareto-like ordering on a set of possible solutions. Then we propose two equivalent ways to recover the result of merging. The first one is based on the notion of compatible rankings defined on finite scales. The second one is only based on total pre-orders induced by ranked bases to merge.
{"title":"A max-based merging of incommensurable ranked belief bases based on finite scales","authors":"S. Benferhat, S. Lagrue, J. Rossit","doi":"10.1109/ICIF.2007.4408136","DOIUrl":"https://doi.org/10.1109/ICIF.2007.4408136","url":null,"abstract":"Recently, several approaches have been proposed to merge possibly contradictory belief bases. This paper focuses on max-based merging operators applied to incommensurable ranked belief bases. We first propose a characterization of a result of merging using Pareto-like ordering on a set of possible solutions. Then we propose two equivalent ways to recover the result of merging. The first one is based on the notion of compatible rankings defined on finite scales. The second one is only based on total pre-orders induced by ranked bases to merge.","PeriodicalId":298941,"journal":{"name":"2007 10th International Conference on Information Fusion","volume":"9 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":"129789769","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.2007.4408027
J. Dolloff, Michelle Iiyama
This paper presents a methodology and supporting fusion algorithm for efficient, sequential, and optimal generation of a ground control network from image block adjustments over an area of interest. Image blocks contain overlapping images (ground footprints) generated from airborne and space-borne sensors, and measurements of ground points in those images. Image block adjustments are ubiquitous in the image geopositioning community and solve for improved image support data (sensor position, attitude, etc.) and geocoordinates of the ground points. The generated ground control network includes the geolocation of the control (aka fiducial and landmark) points and corresponding multi-ground point error covariance or its high-fidelity representation. Experimental results based on simulated data are also presented.
{"title":"Fusion of image block adjustments for the generation of a ground control network","authors":"J. Dolloff, Michelle Iiyama","doi":"10.1109/ICIF.2007.4408027","DOIUrl":"https://doi.org/10.1109/ICIF.2007.4408027","url":null,"abstract":"This paper presents a methodology and supporting fusion algorithm for efficient, sequential, and optimal generation of a ground control network from image block adjustments over an area of interest. Image blocks contain overlapping images (ground footprints) generated from airborne and space-borne sensors, and measurements of ground points in those images. Image block adjustments are ubiquitous in the image geopositioning community and solve for improved image support data (sensor position, attitude, etc.) and geocoordinates of the ground points. The generated ground control network includes the geolocation of the control (aka fiducial and landmark) points and corresponding multi-ground point error covariance or its high-fidelity representation. Experimental results based on simulated data are also presented.","PeriodicalId":298941,"journal":{"name":"2007 10th International Conference on Information Fusion","volume":"8 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":"128422957","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.2007.4408102
Astride Aregui, T. Denoeux
A method is proposed for converting a novelty measure such as produced by one-class SVMs or Kernel principal component analysis (KPCA) into a belief function on a well- defined frame of discernment. This makes it possible to combine one-class classification or novelty detection methods with other information expressed in the same framework such as expert opinions or multi-class classifiers.
{"title":"Fusion of one-class classifiers in the belief function framework","authors":"Astride Aregui, T. Denoeux","doi":"10.1109/ICIF.2007.4408102","DOIUrl":"https://doi.org/10.1109/ICIF.2007.4408102","url":null,"abstract":"A method is proposed for converting a novelty measure such as produced by one-class SVMs or Kernel principal component analysis (KPCA) into a belief function on a well- defined frame of discernment. This makes it possible to combine one-class classification or novelty detection methods with other information expressed in the same framework such as expert opinions or multi-class classifiers.","PeriodicalId":298941,"journal":{"name":"2007 10th International Conference on Information Fusion","volume":"58 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":"130562221","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.2007.4408207
I. Kadar
This paper serves both as introduction to and motivation for the panel, and as a position paper to highlight retrospectives and perspectives on issues and challenges of levels 2/3 fusion implementations by presenting an independent annotated point of view.
{"title":"Results from levels 2/3 fusion implementations: Issues, challenges, retrospectives and perspectives for the future - An annotated view","authors":"I. Kadar","doi":"10.1109/ICIF.2007.4408207","DOIUrl":"https://doi.org/10.1109/ICIF.2007.4408207","url":null,"abstract":"This paper serves both as introduction to and motivation for the panel, and as a position paper to highlight retrospectives and perspectives on issues and challenges of levels 2/3 fusion implementations by presenting an independent annotated point of view.","PeriodicalId":298941,"journal":{"name":"2007 10th International Conference on Information Fusion","volume":"58 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":"127041999","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.2007.4408043
D. Musicki, R. Evans
We investigate the problem of transmitting track state to a fusion center using minimum computational resources. Local track state is not updated and not transmitted, unless local measurements sufficiently decrease local track state uncertainty. Track state information is coded using equivalent innovations. To prevent the accumulation - random walk of coding/decoding errors (coding noise), with potentially unbounded errors, equivalent innovations incorporate the coding noise. Simulations demonstrate negligible information loss when using the equivalent innovations, compared to the optimal central fusion using unlimited bandwidth.
{"title":"Track fusion using equivalent innovations","authors":"D. Musicki, R. Evans","doi":"10.1109/ICIF.2007.4408043","DOIUrl":"https://doi.org/10.1109/ICIF.2007.4408043","url":null,"abstract":"We investigate the problem of transmitting track state to a fusion center using minimum computational resources. Local track state is not updated and not transmitted, unless local measurements sufficiently decrease local track state uncertainty. Track state information is coded using equivalent innovations. To prevent the accumulation - random walk of coding/decoding errors (coding noise), with potentially unbounded errors, equivalent innovations incorporate the coding noise. Simulations demonstrate negligible information loss when using the equivalent innovations, compared to the optimal central fusion using unlimited bandwidth.","PeriodicalId":298941,"journal":{"name":"2007 10th International Conference on Information Fusion","volume":"1 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":"129120591","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.2007.4408130
Marco F. Huber, U. Hanebeck
In nonlinear Bayesian estimation it is generally inevitable to incorporate approximate descriptions of the exact estimation algorithm. There are two possible ways to involve approximations: Approximating the nonlinear stochastic system model or approximating the prior probability density function. The key idea of the introduced novel estimator called Hybrid Density Filter relies on approximating the nonlinear system, thus approximating conditional densities. These densities nonlinearly relate the current system state to the future system state at predictions or to potential measurements at measurement updates. A hybrid density consisting of both Dirac delta functions and Gaussian densities is used for an optimal approximation. This paper addresses the optimization problem for treating the conditional density approximation. Furthermore, efficient estimation algorithms are derived based upon the special structure of the hybrid density, which yield a Gaussian mixture representation of the system state's density.
{"title":"The hybrid density filter for nonlinear estimation based on hybrid conditional density approximation","authors":"Marco F. Huber, U. Hanebeck","doi":"10.1109/ICIF.2007.4408130","DOIUrl":"https://doi.org/10.1109/ICIF.2007.4408130","url":null,"abstract":"In nonlinear Bayesian estimation it is generally inevitable to incorporate approximate descriptions of the exact estimation algorithm. There are two possible ways to involve approximations: Approximating the nonlinear stochastic system model or approximating the prior probability density function. The key idea of the introduced novel estimator called Hybrid Density Filter relies on approximating the nonlinear system, thus approximating conditional densities. These densities nonlinearly relate the current system state to the future system state at predictions or to potential measurements at measurement updates. A hybrid density consisting of both Dirac delta functions and Gaussian densities is used for an optimal approximation. This paper addresses the optimization problem for treating the conditional density approximation. Furthermore, efficient estimation algorithms are derived based upon the special structure of the hybrid density, which yield a Gaussian mixture representation of the system state's density.","PeriodicalId":298941,"journal":{"name":"2007 10th International Conference on Information Fusion","volume":"2004 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":"123906883","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.2007.4408101
J. V. Laere, Maria Nilsson, T. Ziemke
The purpose of information fusion is to support and improve decision making. However, theories of decision making differ significantly on their view of what a good decision actually is. Hence, depending on which decision making theory one (un)consciously adopts there are different requirements for information fusion as decision support Information fusion researchers and practitioners should therefore be more explicit about their assumptions regarding decision making by carefully describing their theoretical frameworks. To illustrate this point the 'theory of sensemaking' by Karl Weick is presented as one example of a decision making theory. Major differences between decision making assumptions in that theory and assumptions common in much information fusion research are highlighted. Implications and challenges for information fusion are discussed.
{"title":"Implications of a Weickian perspective on decision making for information fusion research and practice","authors":"J. V. Laere, Maria Nilsson, T. Ziemke","doi":"10.1109/ICIF.2007.4408101","DOIUrl":"https://doi.org/10.1109/ICIF.2007.4408101","url":null,"abstract":"The purpose of information fusion is to support and improve decision making. However, theories of decision making differ significantly on their view of what a good decision actually is. Hence, depending on which decision making theory one (un)consciously adopts there are different requirements for information fusion as decision support Information fusion researchers and practitioners should therefore be more explicit about their assumptions regarding decision making by carefully describing their theoretical frameworks. To illustrate this point the 'theory of sensemaking' by Karl Weick is presented as one example of a decision making theory. Major differences between decision making assumptions in that theory and assumptions common in much information fusion research are highlighted. Implications and challenges for information fusion are discussed.","PeriodicalId":298941,"journal":{"name":"2007 10th International Conference on Information Fusion","volume":"29 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":"123928976","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：