We present an automatic human skin segmentation algorithm based on color information. The YCbCr color space is employed because it is typically used in video coding and provides an effective use of chrominance information for modeling the human skin color. We model the skin-color distribution as a bivariate normal distribution in the CbCr plane and classify each pixel as skin or non-skin based on its Mahalanobis distance. The performance of the algorithm is illustrated by simulations carried out on images depicting people of different ethnicity.
{"title":"Automatic human skin segmentation based on color information in the YCbCr color space","authors":"N. Habili, C. Lim, A. Moini","doi":"10.1109/IDC.2002.995437","DOIUrl":"https://doi.org/10.1109/IDC.2002.995437","url":null,"abstract":"We present an automatic human skin segmentation algorithm based on color information. The YCbCr color space is employed because it is typically used in video coding and provides an effective use of chrominance information for modeling the human skin color. We model the skin-color distribution as a bivariate normal distribution in the CbCr plane and classify each pixel as skin or non-skin based on its Mahalanobis distance. The performance of the algorithm is illustrated by simulations carried out on images depicting people of different ethnicity.","PeriodicalId":385351,"journal":{"name":"Final Program and Abstracts on Information, Decision and Control","volume":"110 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122746392","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}
A cooperative sensor network is an array of sensors interconnected by a local or wide area communications network. Data is shared between the sensors and used as input to an estimator to measure a process. In the paper a stochastic sensor scheduling framework is applied to the localisation of multiple emitter targets using a cooperative sensor network in which the information rate between sensor nodes has been constrained. The stochastic sensor scheduling problem is presented and a practical sub-optimal sensor scheduling algorithm for multi-target localisation using bearing-only sensors is demonstrated in simulation.
{"title":"Cooperative sensor networks: a stochastic sensor scheduling approach","authors":"L. Sciacca, R. Evans, W. Moran, S. Suvorova","doi":"10.1109/IDC.2002.995394","DOIUrl":"https://doi.org/10.1109/IDC.2002.995394","url":null,"abstract":"A cooperative sensor network is an array of sensors interconnected by a local or wide area communications network. Data is shared between the sensors and used as input to an estimator to measure a process. In the paper a stochastic sensor scheduling framework is applied to the localisation of multiple emitter targets using a cooperative sensor network in which the information rate between sensor nodes has been constrained. The stochastic sensor scheduling problem is presented and a practical sub-optimal sensor scheduling algorithm for multi-target localisation using bearing-only sensors is demonstrated in simulation.","PeriodicalId":385351,"journal":{"name":"Final Program and Abstracts on Information, Decision and Control","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125243473","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}
We present three related algorithms for iteratively computing all the eigenvectors of a Hermitian matrix. The algorithms are based on the idea of applying Newton updates to individual eigenvectors at each iteration. The advantage of these Newton updates is that they have a cubic rate of convergence. The difference between the algorithms is how they prevent the individual updates from converging to the same eigenvector. The first algorithm finds the eigenvectors sequentially, and uses a novel form of deflation in order to ensure all the eigenvectors are found. Rather than modify the matrix directly, which introduces large errors if the matrix is ill-conditioned, deflation is achieved by restricting the Newton updates to lie in a subspace orthogonal to all previously found eigenvectors. The other algorithms estimate all the eigenvectors at once. At each iteration, they sweep through all the estimates, performing a Newton update on each estimate once per sweep. Orthogonality is maintained by explicit re-orthogonalisation after each update, which also serves to improve the asymptotic rate of convergence of the algorithms.
{"title":"Novel Newton algorithms for the Hermitian eigenvalue problem","authors":"M. Nikpour, J. Manton, R. Mahony","doi":"10.1109/IDC.2002.995439","DOIUrl":"https://doi.org/10.1109/IDC.2002.995439","url":null,"abstract":"We present three related algorithms for iteratively computing all the eigenvectors of a Hermitian matrix. The algorithms are based on the idea of applying Newton updates to individual eigenvectors at each iteration. The advantage of these Newton updates is that they have a cubic rate of convergence. The difference between the algorithms is how they prevent the individual updates from converging to the same eigenvector. The first algorithm finds the eigenvectors sequentially, and uses a novel form of deflation in order to ensure all the eigenvectors are found. Rather than modify the matrix directly, which introduces large errors if the matrix is ill-conditioned, deflation is achieved by restricting the Newton updates to lie in a subspace orthogonal to all previously found eigenvectors. The other algorithms estimate all the eigenvectors at once. At each iteration, they sweep through all the estimates, performing a Newton update on each estimate once per sweep. Orthogonality is maintained by explicit re-orthogonalisation after each update, which also serves to improve the asymptotic rate of convergence of the algorithms.","PeriodicalId":385351,"journal":{"name":"Final Program and Abstracts on Information, Decision and Control","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131412669","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}
An important problem in multi-target tracking in clutter is that of track initiation and termination. When targets enter or leave the field of view of a sensor, the tracking algorithm must start new tracks or terminate existing tracks accordingly. One approach used to solve this problem is the addition of a visibility attribute to tracks. The visibility attribute provides a degree of filtering over the prior probability of measurement association and can be used as a statistic for track decisions. Standard model order estimation techniques suggest an alternative approach based on the log likelihood of the measurements. These approaches are used to derive initiation schemes for the probabilistic multi-hypothesis tracker (PMHT). The performance of these two approaches is compared with an intuitive ad hoc system through the use of receiver operating characteristic type curves.
{"title":"A comparison of track initiation methods with the PMHT","authors":"S. Davey, D. Gray","doi":"10.1109/IDC.2002.995423","DOIUrl":"https://doi.org/10.1109/IDC.2002.995423","url":null,"abstract":"An important problem in multi-target tracking in clutter is that of track initiation and termination. When targets enter or leave the field of view of a sensor, the tracking algorithm must start new tracks or terminate existing tracks accordingly. One approach used to solve this problem is the addition of a visibility attribute to tracks. The visibility attribute provides a degree of filtering over the prior probability of measurement association and can be used as a statistic for track decisions. Standard model order estimation techniques suggest an alternative approach based on the log likelihood of the measurements. These approaches are used to derive initiation schemes for the probabilistic multi-hypothesis tracker (PMHT). The performance of these two approaches is compared with an intuitive ad hoc system through the use of receiver operating characteristic type curves.","PeriodicalId":385351,"journal":{"name":"Final Program and Abstracts on Information, Decision and Control","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131845718","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}
An overview of a system for the automatic classification of aircraft from flyover data is presented. The system is passive, utilising acoustic sensors to measure both broadband and narrowband energy. Aspects of the system architecture, sensor design and signal processing are covered. The processing is divided into three streams: broadband, narrowband and cepstrum. Each processing stream is capable of extracting flight parameter estimates from the acoustic data. Broadband estimation is based on the time-delay cross correlation of signals from multiple sensors. Narrowband estimation makes use of the spectrogram of the data to extract frequency lines produced by the aircraft and subject to the acoustical Doppler effect. Cepstrum processing tracks the primary rahmonic in the cepstrogram due to multipath interference. A novel hidden Markov model tracking technique is applied to form tracks on the noisy spectrogram and cepstrogram data. Examples of real data processing and flight parameter estimates for classification are given.
{"title":"A system for automatic classification of aircraft flyovers using acoustic data","authors":"J. Sendt, G. Pulford, Yujin Gao, A. Maguer","doi":"10.1109/IDC.2002.995387","DOIUrl":"https://doi.org/10.1109/IDC.2002.995387","url":null,"abstract":"An overview of a system for the automatic classification of aircraft from flyover data is presented. The system is passive, utilising acoustic sensors to measure both broadband and narrowband energy. Aspects of the system architecture, sensor design and signal processing are covered. The processing is divided into three streams: broadband, narrowband and cepstrum. Each processing stream is capable of extracting flight parameter estimates from the acoustic data. Broadband estimation is based on the time-delay cross correlation of signals from multiple sensors. Narrowband estimation makes use of the spectrogram of the data to extract frequency lines produced by the aircraft and subject to the acoustical Doppler effect. Cepstrum processing tracks the primary rahmonic in the cepstrogram due to multipath interference. A novel hidden Markov model tracking technique is applied to form tracks on the noisy spectrogram and cepstrogram data. Examples of real data processing and flight parameter estimates for classification are given.","PeriodicalId":385351,"journal":{"name":"Final Program and Abstracts on Information, Decision and Control","volume":"28 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130361388","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}
There are different ways of assessing the intent of a target. However, the most effective way is one that can offer the most accurate estimates compared to human-defined benchmarks. To allow various intent assessment technologies to be evaluated against the benchmarks, a reconfigurable simulation environment has been designed using a sound software engineering process which is the Rational Unified Process and the Unified Modelling Language. In order to cope with future tests with different technologies, the software architecture was structured in such a way that the simulation environment talks to different intent assessment technologies via an interface, hence forming a plug-and-play simulation environment.
{"title":"Automated intent assessment simulation environment","authors":"X. Nguyen, P. Heuer","doi":"10.1109/IDC.2002.995440","DOIUrl":"https://doi.org/10.1109/IDC.2002.995440","url":null,"abstract":"There are different ways of assessing the intent of a target. However, the most effective way is one that can offer the most accurate estimates compared to human-defined benchmarks. To allow various intent assessment technologies to be evaluated against the benchmarks, a reconfigurable simulation environment has been designed using a sound software engineering process which is the Rational Unified Process and the Unified Modelling Language. In order to cope with future tests with different technologies, the software architecture was structured in such a way that the simulation environment talks to different intent assessment technologies via an interface, hence forming a plug-and-play simulation environment.","PeriodicalId":385351,"journal":{"name":"Final Program and Abstracts on Information, Decision and Control","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116818673","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}
Discrete frequency domain adaptive filtering is often used for the restoration of signals in the presence of interference. Often, this is achieved by an approximate Wiener filter implementation. Sometimes, a desired response signal is derived from the received signal and so may be itself, contaminated by interference. If an interferer is narrowband and approximately bin centred, it constitutes a DC offset within that bin. This can reduce the quality of a Wiener filter restoration of the signal, irrespective of whether the interference contaminates either the primary, reference, or both signals. However, the DC offset can be readily removed using a simple technique. We outline this technique and demonstrate the considerable performance improvement through examples.
{"title":"Improved narrowband interference rejection using frequency domain DC offset compensation","authors":"G. Parker, K. Lever","doi":"10.1109/IDC.2002.995388","DOIUrl":"https://doi.org/10.1109/IDC.2002.995388","url":null,"abstract":"Discrete frequency domain adaptive filtering is often used for the restoration of signals in the presence of interference. Often, this is achieved by an approximate Wiener filter implementation. Sometimes, a desired response signal is derived from the received signal and so may be itself, contaminated by interference. If an interferer is narrowband and approximately bin centred, it constitutes a DC offset within that bin. This can reduce the quality of a Wiener filter restoration of the signal, irrespective of whether the interference contaminates either the primary, reference, or both signals. However, the DC offset can be readily removed using a simple technique. We outline this technique and demonstrate the considerable performance improvement through examples.","PeriodicalId":385351,"journal":{"name":"Final Program and Abstracts on Information, Decision and Control","volume":"117 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123446337","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}
This paper introduces an application of the modern robust control theory to the missile guidance problem where the successful intercept criterion has been defined in terms of minimizing the terminal miss distance. This output feedback controller steers the initial state of the missile system into an ellipsoidal bounding set in a specified flight time. The underlying system in this problem is a time-varying uncertain linear system in which the uncertainty satisfies a certain integral quadratic constraint. The performance of the robust controller was compared with output LQR feedback controller and proportional navigation (PN) controllers.
{"title":"Robust control applied to missile guidance","authors":"P. N. Pathirana, A. Savkin","doi":"10.1109/IDC.2002.995385","DOIUrl":"https://doi.org/10.1109/IDC.2002.995385","url":null,"abstract":"This paper introduces an application of the modern robust control theory to the missile guidance problem where the successful intercept criterion has been defined in terms of minimizing the terminal miss distance. This output feedback controller steers the initial state of the missile system into an ellipsoidal bounding set in a specified flight time. The underlying system in this problem is a time-varying uncertain linear system in which the uncertainty satisfies a certain integral quadratic constraint. The performance of the robust controller was compared with output LQR feedback controller and proportional navigation (PN) controllers.","PeriodicalId":385351,"journal":{"name":"Final Program and Abstracts on Information, Decision and Control","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125605238","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}
The acoustic signatures of various types of ground vehicles are recorded during each vehicle's transit past a stationary microphone located just above ground level and the acoustic data are analysed using the short time Fourier transform. The spectrogram for each transiting vehicle exhibits a primary set and a secondary set of harmonic lines in the time-frequency domain that can be used for classification purposes. The frequency of each harmonic line varies with time due to the acoustical Doppler effect. The variation with time of the instantaneous frequency of the strongest harmonic line is used to estimate the speed, ground distance at the closest point of approach to the microphone, and engine-firing rate of the vehicle.
{"title":"Time frequency signal analysis of ground vehicle acoustic signatures","authors":"K. Lo, B. Ferguson","doi":"10.1109/IDC.2002.995415","DOIUrl":"https://doi.org/10.1109/IDC.2002.995415","url":null,"abstract":"The acoustic signatures of various types of ground vehicles are recorded during each vehicle's transit past a stationary microphone located just above ground level and the acoustic data are analysed using the short time Fourier transform. The spectrogram for each transiting vehicle exhibits a primary set and a secondary set of harmonic lines in the time-frequency domain that can be used for classification purposes. The frequency of each harmonic line varies with time due to the acoustical Doppler effect. The variation with time of the instantaneous frequency of the strongest harmonic line is used to estimate the speed, ground distance at the closest point of approach to the microphone, and engine-firing rate of the vehicle.","PeriodicalId":385351,"journal":{"name":"Final Program and Abstracts on Information, Decision and Control","volume":"2016 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125719142","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}
Video motion detection is fundamental in many autonomous video surveillance strategies. However, in outdoor scenes where inconsistent lighting and unimportant, but distracting, background movement is present, it is a challenging problem. Recent research has produced several background modelling techniques, based on image differencing, that exhibit real-time performance and high accuracy for certain classes of scene. The aim of this paper is to assess the performance of some of these background modelling techniques, namely the Gaussian mixture model and the hybrid detection algorithm, using video sequences of outdoor scenes where the weather introduces unpredictable variations in both lighting and background movement. The results are analysed and reported, with the aim of identifying suitable directions for enhancing the robustness of motion detection techniques for outdoor video surveillance systems.
{"title":"Evaluation of motion detection techniques for video surveillance","authors":"M. Fettke, K. Sammut, M. Naylor, F. He","doi":"10.1109/IDC.2002.995405","DOIUrl":"https://doi.org/10.1109/IDC.2002.995405","url":null,"abstract":"Video motion detection is fundamental in many autonomous video surveillance strategies. However, in outdoor scenes where inconsistent lighting and unimportant, but distracting, background movement is present, it is a challenging problem. Recent research has produced several background modelling techniques, based on image differencing, that exhibit real-time performance and high accuracy for certain classes of scene. The aim of this paper is to assess the performance of some of these background modelling techniques, namely the Gaussian mixture model and the hybrid detection algorithm, using video sequences of outdoor scenes where the weather introduces unpredictable variations in both lighting and background movement. The results are analysed and reported, with the aim of identifying suitable directions for enhancing the robustness of motion detection techniques for outdoor video surveillance systems.","PeriodicalId":385351,"journal":{"name":"Final Program and Abstracts on Information, Decision and Control","volume":"83 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121393978","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：