Evaluating Signature Matching in a Multi-Sensor Vehicle Classification System (extended) . Zhang, C. & Heidemann, J. Technical Report ISI-TR-2011-675, USC/Information Sciences Institute, November, 2011. ![Evaluating Signature Matching in a Multi-Sensor Vehicle Classification System (extended) [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper abstract bibtex Many academic sensornet systems consider the problem of tracking targets as they move through a fixed field of sensors. Such applications must relate sensor \emphdetections to specific \emphtargets, yet prior work has often ignored this problem, assuming either a single target, sufficient spatial separation that target-to-sensor mapping is clear, or some out-of-band detection-to-target mapping. There has been little study of algorithms to associate detections with targets, and effects of their accuracy on detection results, particularly in environments with dense targets. We explore the question of \emphdetection-to-target mapping in the context of a \emphvehicle classification system for urban roadways, where vehicles pass fixed sensors at varying but frequent rates. We develop several \emphsignature matching algorithms that relate detections at different sensors to same or different vehicles. We evaluate these algorithms with data taken in a field test of live traffic compared against ground truth obtained through manual analysis of video and the resulting matching recall is over 78%. We investigate the effects of mapping accuracy on length-based vehicle classification. We show that accurate signature matching is critical to multi-sensor algorithms. We compare our matching algorithms against an oracle (perfect information), and find that all matching reduces end-to-end accuracy somewhat, but a poor matching algorithm reduce accuracy by 21%, while our best algorithm reduces it by only 10% in our case study. Finally, we quantify the degree of correlation between matching correctness and classification accuracy.
@TechReport{Zhang11c,
author = "Chengjie Zhang and John Heidemann",
title = "Evaluating Signature Matching in a Multi-Sensor Vehicle Classification System (extended) ",
institution = "USC/Information Sciences Institute",
year = 2011,
sortdate = "2011-11-01",
project = "ilense, cisoft, surese",
jsubject = "sensornet_fusion",
number = "ISI-TR-2011-675",
month = nov,
location = "johnh: pafile",
keywords = "signature matching, multi-sensor",
url = "http://www.isi.edu/%7ejohnh/PAPERS/Zhang11c.html",
pdfurl = "http://www.isi.edu/%7ejohnh/PAPERS/Zhang11c.pdf",
myorganization = "USC/Information Sciences Institute",
copyrightholder = "authors",
abstract = "
Many academic sensornet systems consider the problem of tracking
targets as they move through a fixed field of sensors. Such
applications must relate sensor \emph{detections} to
specific \emph{targets}, yet prior work has often ignored this problem,
assuming either a single target, sufficient spatial separation that
target-to-sensor mapping is clear, or some out-of-band
detection-to-target mapping. There has been little study of
algorithms to associate detections with targets, and effects of their
accuracy on detection results, particularly in environments with dense
targets. We explore the question of \emph{detection-to-target}
mapping in the context of a \emph{vehicle classification system} for
urban roadways, where vehicles pass fixed sensors at varying but
frequent rates. We develop several \emph{signature matching}
algorithms that relate detections at different sensors to same or
different vehicles. We evaluate these algorithms with data taken in a
field test of live traffic compared against ground truth obtained
through manual analysis of video and the resulting matching recall is
over 78\%. We investigate the effects of mapping accuracy on
length-based vehicle classification. We show that accurate signature
matching is critical to multi-sensor algorithms. We compare our
matching algorithms against an oracle (perfect information), and find
that all matching reduces end-to-end accuracy somewhat, but a poor
matching algorithm reduce accuracy by 21\%, while our best algorithm
reduces it by only 10\% in our case study. Finally, we quantify the
degree of correlation between matching correctness and classification
accuracy.
",
}
Downloads: 0
{"_id":"yJDcTKYDmqRoWuvDA","bibbaseid":"zhang-heidemann-evaluatingsignaturematchinginamultisensorvehicleclassificationsystemextended-2011","author_short":["Zhang, C.","Heidemann, J."],"bibdata":{"bibtype":"techreport","type":"techreport","author":[{"firstnames":["Chengjie"],"propositions":[],"lastnames":["Zhang"],"suffixes":[]},{"firstnames":["John"],"propositions":[],"lastnames":["Heidemann"],"suffixes":[]}],"title":"Evaluating Signature Matching in a Multi-Sensor Vehicle Classification System (extended) ","institution":"USC/Information Sciences Institute","year":"2011","sortdate":"2011-11-01","project":"ilense, cisoft, surese","jsubject":"sensornet_fusion","number":"ISI-TR-2011-675","month":"November","location":"johnh: pafile","keywords":"signature matching, multi-sensor","url":"http://www.isi.edu/%7ejohnh/PAPERS/Zhang11c.html","pdfurl":"http://www.isi.edu/%7ejohnh/PAPERS/Zhang11c.pdf","myorganization":"USC/Information Sciences Institute","copyrightholder":"authors","abstract":"Many academic sensornet systems consider the problem of tracking targets as they move through a fixed field of sensors. Such applications must relate sensor \\emphdetections to specific \\emphtargets, yet prior work has often ignored this problem, assuming either a single target, sufficient spatial separation that target-to-sensor mapping is clear, or some out-of-band detection-to-target mapping. There has been little study of algorithms to associate detections with targets, and effects of their accuracy on detection results, particularly in environments with dense targets. We explore the question of \\emphdetection-to-target mapping in the context of a \\emphvehicle classification system for urban roadways, where vehicles pass fixed sensors at varying but frequent rates. We develop several \\emphsignature matching algorithms that relate detections at different sensors to same or different vehicles. We evaluate these algorithms with data taken in a field test of live traffic compared against ground truth obtained through manual analysis of video and the resulting matching recall is over 78%. We investigate the effects of mapping accuracy on length-based vehicle classification. We show that accurate signature matching is critical to multi-sensor algorithms. We compare our matching algorithms against an oracle (perfect information), and find that all matching reduces end-to-end accuracy somewhat, but a poor matching algorithm reduce accuracy by 21%, while our best algorithm reduces it by only 10% in our case study. Finally, we quantify the degree of correlation between matching correctness and classification accuracy. ","bibtex":"@TechReport{Zhang11c,\n\tauthor = \t\"Chengjie Zhang and John Heidemann\",\n\ttitle = \t\"Evaluating Signature Matching in a Multi-Sensor Vehicle Classification System (extended) \",\n\tinstitution = \t\"USC/Information Sciences Institute\",\n\tyear = \t\t2011,\n\tsortdate = \"2011-11-01\",\n\tproject = \"ilense, cisoft, surese\",\n\tjsubject = \"sensornet_fusion\",\n\tnumber =\t\"ISI-TR-2011-675\",\n\tmonth =\t\tnov,\n\tlocation =\t\"johnh: pafile\",\n\tkeywords =\t\"signature matching, multi-sensor\",\n\turl =\t\t\"http://www.isi.edu/%7ejohnh/PAPERS/Zhang11c.html\",\n\tpdfurl =\t\"http://www.isi.edu/%7ejohnh/PAPERS/Zhang11c.pdf\",\n\tmyorganization =\t\"USC/Information Sciences Institute\",\n\tcopyrightholder = \"authors\",\n\tabstract = \"\nMany academic sensornet systems consider the problem of tracking\ntargets as they move through a fixed field of sensors. Such\napplications must relate sensor \\emph{detections} to \nspecific \\emph{targets}, yet prior work has often ignored this problem,\nassuming either a single target, sufficient spatial separation that\ntarget-to-sensor mapping is clear, or some out-of-band\ndetection-to-target mapping. There has been little study of\nalgorithms to associate detections with targets, and effects of their\naccuracy on detection results, particularly in environments with dense\ntargets. We explore the question of \\emph{detection-to-target}\nmapping in the context of a \\emph{vehicle classification system} for\nurban roadways, where vehicles pass fixed sensors at varying but\nfrequent rates. We develop several \\emph{signature matching}\nalgorithms that relate detections at different sensors to same or\ndifferent vehicles. We evaluate these algorithms with data taken in a\nfield test of live traffic compared against ground truth obtained\nthrough manual analysis of video and the resulting matching recall is\nover 78\\%. We investigate the effects of mapping accuracy on\nlength-based vehicle classification. We show that accurate signature\nmatching is critical to multi-sensor algorithms. We compare our\nmatching algorithms against an oracle (perfect information), and find\nthat all matching reduces end-to-end accuracy somewhat, but a poor\nmatching algorithm reduce accuracy by 21\\%, while our best algorithm\nreduces it by only 10\\% in our case study. Finally, we quantify the\ndegree of correlation between matching correctness and classification\naccuracy.\n\",\n}\n\n","author_short":["Zhang, C.","Heidemann, J."],"bibbaseid":"zhang-heidemann-evaluatingsignaturematchinginamultisensorvehicleclassificationsystemextended-2011","role":"author","urls":{"Paper":"http://www.isi.edu/%7ejohnh/PAPERS/Zhang11c.html"},"keyword":["signature matching","multi-sensor"],"metadata":{"authorlinks":{}}},"bibtype":"techreport","biburl":"https://bibbase.org/f/dHevizJoWEhWowz8q/johnh-2023-2.bib","dataSources":["YLyu3mj3xsBeoqiHK","fLZcDgNSoSuatv6aX","fxEParwu2ZfurScPY","7nuQvtHTqKrLmgu99"],"keywords":["signature matching","multi-sensor"],"search_terms":["evaluating","signature","matching","multi","sensor","vehicle","classification","system","extended","zhang","heidemann"],"title":"Evaluating Signature Matching in a Multi-Sensor Vehicle Classification System (extended) ","year":2011}