(Georgia Institute of Technology, 2002)
Wardi, Yorai Y.; Riley, George F.
This paper considers congestion-related performance metrics in tandem networks of Stochastic Fluid Models
(SFMs), and derives their IPA gradient estimators with
respect to buffer sizes. Specifically, the performance metrics
in question are the total loss volume and the cumulative
buffer workload (buffer contents), and the control
parameter consists of buffer limits at both the node
where the performance is measured and at an upstream
node. The IPA estimators are unbiased and nonparametric,
and hence can be computed on-line from field measurements
as well as off-line from simulation experiments.
The IPA derivatives are applied to packet-based networks,
where simulation results support the theoretical developments.
Possible applications to congestion management in telecommunications networks are discussed.
(Georgia Institute of Technology, 2002)
Ding, Yuhua; Vachtsevanos, George J.; Yezzi, Anthony; Zhang, Yingchuan; Wardi, Yorai Y.
Segmentation accuracy is shown to be a critical factor in
detection rate improvement. With accurate segmentation,
results are easier to interpret, and classification
performance is better. Therefore, it is required to have a
performance measure for segmentation evaluation.
However, a number of restrictions limit using existing
segmentation performance measures. In this paper a recursive segmentation and classification scheme is proposed to improve segmentation accuracy and
classification performance in real-time machine vision applications. In this scheme, the confidence level of
classification results is used as a new performance
measure to evaluate the accuracy of segmentation
algorithm. Segmentation is repeated until a classification
with desired confidence level is achieved. This scheme can be implemented automatically. Experimental results
show that it is efficient to improve segmentation accuracy
and the overall detection performance, especially for
real-time machine vision applications, where the scene is
complicated and a single segmentation algorithm cannot
produce satisfactory results.