An online kernel change detection algorithm. Desobry, F., Davy, M., & Doncarli, C. IEEE Transactions on Signal Processing, 53(8):2961–2974, August, 2005. Conference Name: IEEE Transactions on Signal Processingdoi abstract bibtex A number of abrupt change detection methods have been proposed in the past, among which are efficient model-based techniques such as the Generalized Likelihood Ratio (GLR) test. We consider the case where no accurate nor tractable model can be found, using a model-free approach, called Kernel change detection (KCD). KCD compares two sets of descriptors extracted online from the signal at each time instant: The immediate past set and the immediate future set. Based on the soft margin single-class Support Vector Machine (SVM), we build a dissimilarity measure in feature space between those sets, without estimating densities as an intermediary step. This dissimilarity measure is shown to be asymptotically equivalent to the Fisher ratio in the Gaussian case. Implementation issues are addressed; in particular, the dissimilarity measure can be computed online in input space. Simulation results on both synthetic signals and real music signals show the efficiency of KCD.
@article{desobry_online_2005,
title = {An online kernel change detection algorithm},
volume = {53},
issn = {1941-0476},
doi = {10.1109/TSP.2005.851098},
abstract = {A number of abrupt change detection methods have been proposed in the past, among which are efficient model-based techniques such as the Generalized Likelihood Ratio (GLR) test. We consider the case where no accurate nor tractable model can be found, using a model-free approach, called Kernel change detection (KCD). KCD compares two sets of descriptors extracted online from the signal at each time instant: The immediate past set and the immediate future set. Based on the soft margin single-class Support Vector Machine (SVM), we build a dissimilarity measure in feature space between those sets, without estimating densities as an intermediary step. This dissimilarity measure is shown to be asymptotically equivalent to the Fisher ratio in the Gaussian case. Implementation issues are addressed; in particular, the dissimilarity measure can be computed online in input space. Simulation results on both synthetic signals and real music signals show the efficiency of KCD.},
number = {8},
journal = {IEEE Transactions on Signal Processing},
author = {Desobry, F. and Davy, M. and Doncarli, C.},
month = aug,
year = {2005},
note = {Conference Name: IEEE Transactions on Signal Processing},
keywords = {Abrupt change detection, Density measurement, Detection algorithms, Extraterrestrial measurements, Kernel, Multiple signal classification, Object detection, Particle measurements, Signal processing, Support vector machines, Testing, kernel method, music segmentation, online, single-class SVM},
pages = {2961--2974},
}
Downloads: 0
{"_id":"LqvHn5KpQsyKMp4Ax","bibbaseid":"desobry-davy-doncarli-anonlinekernelchangedetectionalgorithm-2005","downloads":0,"creationDate":"2018-01-22T16:01:03.186Z","title":"An online kernel change detection algorithm","author_short":["Desobry, F.","Davy, M.","Doncarli, C."],"year":2005,"bibtype":"article","biburl":"https://bibbase.org/zotero/mh_lenguyen","bibdata":{"bibtype":"article","type":"article","title":"An online kernel change detection algorithm","volume":"53","issn":"1941-0476","doi":"10.1109/TSP.2005.851098","abstract":"A number of abrupt change detection methods have been proposed in the past, among which are efficient model-based techniques such as the Generalized Likelihood Ratio (GLR) test. We consider the case where no accurate nor tractable model can be found, using a model-free approach, called Kernel change detection (KCD). KCD compares two sets of descriptors extracted online from the signal at each time instant: The immediate past set and the immediate future set. Based on the soft margin single-class Support Vector Machine (SVM), we build a dissimilarity measure in feature space between those sets, without estimating densities as an intermediary step. This dissimilarity measure is shown to be asymptotically equivalent to the Fisher ratio in the Gaussian case. Implementation issues are addressed; in particular, the dissimilarity measure can be computed online in input space. Simulation results on both synthetic signals and real music signals show the efficiency of KCD.","number":"8","journal":"IEEE Transactions on Signal Processing","author":[{"propositions":[],"lastnames":["Desobry"],"firstnames":["F."],"suffixes":[]},{"propositions":[],"lastnames":["Davy"],"firstnames":["M."],"suffixes":[]},{"propositions":[],"lastnames":["Doncarli"],"firstnames":["C."],"suffixes":[]}],"month":"August","year":"2005","note":"Conference Name: IEEE Transactions on Signal Processing","keywords":"Abrupt change detection, Density measurement, Detection algorithms, Extraterrestrial measurements, Kernel, Multiple signal classification, Object detection, Particle measurements, Signal processing, Support vector machines, Testing, kernel method, music segmentation, online, single-class SVM","pages":"2961–2974","bibtex":"@article{desobry_online_2005,\n\ttitle = {An online kernel change detection algorithm},\n\tvolume = {53},\n\tissn = {1941-0476},\n\tdoi = {10.1109/TSP.2005.851098},\n\tabstract = {A number of abrupt change detection methods have been proposed in the past, among which are efficient model-based techniques such as the Generalized Likelihood Ratio (GLR) test. We consider the case where no accurate nor tractable model can be found, using a model-free approach, called Kernel change detection (KCD). KCD compares two sets of descriptors extracted online from the signal at each time instant: The immediate past set and the immediate future set. Based on the soft margin single-class Support Vector Machine (SVM), we build a dissimilarity measure in feature space between those sets, without estimating densities as an intermediary step. This dissimilarity measure is shown to be asymptotically equivalent to the Fisher ratio in the Gaussian case. Implementation issues are addressed; in particular, the dissimilarity measure can be computed online in input space. Simulation results on both synthetic signals and real music signals show the efficiency of KCD.},\n\tnumber = {8},\n\tjournal = {IEEE Transactions on Signal Processing},\n\tauthor = {Desobry, F. and Davy, M. and Doncarli, C.},\n\tmonth = aug,\n\tyear = {2005},\n\tnote = {Conference Name: IEEE Transactions on Signal Processing},\n\tkeywords = {Abrupt change detection, Density measurement, Detection algorithms, Extraterrestrial measurements, Kernel, Multiple signal classification, Object detection, Particle measurements, Signal processing, Support vector machines, Testing, kernel method, music segmentation, online, single-class SVM},\n\tpages = {2961--2974},\n}\n\n\n\n","author_short":["Desobry, F.","Davy, M.","Doncarli, C."],"key":"desobry_online_2005","id":"desobry_online_2005","bibbaseid":"desobry-davy-doncarli-anonlinekernelchangedetectionalgorithm-2005","role":"author","urls":{},"keyword":["Abrupt change detection","Density measurement","Detection algorithms","Extraterrestrial measurements","Kernel","Multiple signal classification","Object detection","Particle measurements","Signal processing","Support vector machines","Testing","kernel method","music segmentation","online","single-class SVM"],"metadata":{"authorlinks":{}},"downloads":0,"html":""},"search_terms":["online","kernel","change","detection","algorithm","desobry","davy","doncarli"],"keywords":["abrupt change detection","density measurement","detection algorithms","extraterrestrial measurements","kernel","multiple signal classification","object detection","particle measurements","signal processing","support vector machines","testing","kernel method","music segmentation","online","single-class svm"],"authorIDs":[],"dataSources":["9cexBw6hrwgyZphZZ","iwKepCrWBps7ojhDx"]}