Electrical Engineering
      and Computer Sciences

Electrical Engineering and Computer Sciences

COLLEGE OF ENGINEERING

UC Berkeley

Estimating Data Stream Quality for Object-Detection Applications

Anish Das Sarma, Shawn Ryan Jeffery, Michael Franklin and Jennifer Widom

EECS Department
University of California, Berkeley
Technical Report No. UCB/EECS-2005-23
December 6, 2005

http://www.eecs.berkeley.edu/Pubs/TechRpts/2005/EECS-2005-23.pdf

Object-detection applications rely on streams of data gathered from sensors, RFID readers, and image recognition systems, among others. These raw data streams tend to be noisy, including both false positives (erroneous readings) and false negatives (missed readings). Techniques exist for general-purpose cleaning of these types of data streams, based on temporal and/or spatial correlations, as well as properties of the physical world. Cleaning is effective at improving the quality of the data, however no cleaning procedures can eliminate all errors. In this paper we identify and address the problem of quality estimation as object-detection data streams are cleaned. We provide techniques for estimating both confidence and coverage s streams are processed by cleaning modules. Detailed experimental results based on an RFID application demonstrate the accuracy and effectiveness of our approach.


BibTeX citation:

@techreport{Das Sarma:EECS-2005-23,
    Author = {Das Sarma, Anish and Jeffery, Shawn Ryan and Franklin, Michael and Widom, Jennifer},
    Title = {Estimating Data Stream Quality for Object-Detection Applications},
    Institution = {EECS Department, University of California, Berkeley},
    Year = {2005},
    Month = {Dec},
    URL = {http://www.eecs.berkeley.edu/Pubs/TechRpts/2005/EECS-2005-23.html},
    Number = {UCB/EECS-2005-23},
    Abstract = {Object-detection applications rely on streams of data gathered from sensors, RFID readers, and image recognition systems, among others. These raw data streams tend to be noisy, including both false positives (erroneous readings) and false negatives (missed readings). Techniques exist for general-purpose cleaning of these types of data streams, based on temporal and/or spatial correlations, as well as properties of the physical world. Cleaning is effective at improving the quality of the data, however no cleaning procedures can eliminate all errors. In this paper we identify and address the problem of quality estimation as object-detection data streams are cleaned. We provide techniques for estimating both confidence and coverage s streams are processed by cleaning modules. Detailed experimental results based on an RFID application demonstrate the accuracy and effectiveness of our approach.}
}

EndNote citation:

%0 Report
%A Das Sarma, Anish
%A Jeffery, Shawn Ryan
%A Franklin, Michael
%A Widom, Jennifer
%T Estimating Data Stream Quality for Object-Detection Applications
%I EECS Department, University of California, Berkeley
%D 2005
%8 December 6
%@ UCB/EECS-2005-23
%U http://www.eecs.berkeley.edu/Pubs/TechRpts/2005/EECS-2005-23.html
%F Das Sarma:EECS-2005-23