\documentclass[11pt]{article}
\setlength{\oddsidemargin}{0.0truein}
\setlength{\evensidemargin}{0.0truein}
\setlength{\textwidth}{6.5truein}
\setlength{\topmargin}{0.0truein}
\setlength{\textheight}{9.0truein} \setlength{\headsep}{0.0truein}
\setlength{\headheight}{0.0truein} \setlength{\topskip}{10.0pt}
\setlength{\parskip}{5mm}
\usepackage{url}
\usepackage{amsmath}
\usepackage{amssymb}
\begin{document}

\begin{center}
  \textbf{\textsc{STANFORD UNIVERSITY}}\\[5pt]
  \textbf{\textsc{DEPARTMENT OF STATISTICS}}\\[5pt]
  \Large{\textbf\textsc{{DEPARTMENTAL SEMINAR}}}
\end{center}
\begin{center}
  4:15 p.m., Tuesday, November 21, 2006\\
  Sequoia Hall Room 200\\
  (Cookies at 3:45 in 1st Floor Lounge)
\end{center}


\begin{center}
  \textsl{Bin Yu} \\
Department of Statistics\\
University of California, Berkeley
\end{center}
\begin{center}
  \textbf{ Arctic Cloud Detection using Multi-Angle and Hyperspectral Satellite Images}
\end{center}

\noindent                      

Sensitivity of Earth's climate to increasing amounts of atmospheric 
carbon dioxide is a topic of general and scientific interest and 
public policy.  Today's global climate models predict
that the strongest dependences of surface temperatures on increasing 
atmospheric carbon dioxide levels will occur in the Arctic.  
Ascertaining the properties of clouds in the Arctic 
via conventional satelliate images is a challenging problem because
liquid and ice water cloud particles often have similar properties to 
the snow and ice particles that compose snow- and ice-covered surfaces. 
Without accurate characterization of clouds over the Arctic we will 
not be able to assess the impact of clouds on the flow of solar and
terrestrial electromagnetic radiation through the Arctic atmosphere and 
we will not be able to ascertain whether they are changing in ways 
that enhance or ameliorate future warming in the Arctic.
With the launch of the Multi-angle Imaging SpectroRadiometer (MISR) and 
Moderate Resolution Imaging Spectrometer (MODIS) by NASA in 1999,
novel electromagnetic radiation measurements made at many angles 
(from MISR) and across many narrow wavelength (hyperspectral)
regions in the visible and infrared (by MODIS) became available for 
scientific study.

In this talk, we report on an on-going collaborative effort 
since 2001 to devise online and effective arctic cloud detection 
algorithms based on both MISR and MODIS measurements.
After much interaction with the MISR team at JPL, we have arrived
at three physically meaningful features from MISR for our clustering 
algorithm, Enhanced Linear Correlation Matching Clustering 
(ELCMC). This algorithm thresholds the three features
with two fixed thresholds and one adaptive threshold which
is found by an EM algorithm. It is robust and computationally fast
for online processing of MISR images.  Further improvements over 
ELCMC are achieved by using concensus lables of ELCMC-MISR and 
MODIS to train Fisher's Quadratic Discriminat Analysis (QDA)
to provide soft labels.  Support vector machines (SVMs) can obtain 
better accuracies when compared with expert labels for some images 
than QDA, but they are much slower than QDA. Binning is then applied 
to the features before SVMs to speed up the SVM computation 
substantially with comparable cloud detection results.  

The cloud labels from different algorithms are compared with the 
best "ground truth" available in large quantities -- the expert labels.
QDA based on the consensus labels from MISR-ELCMC and MODIS
achieves a 94% accuracy and a 100% coverage (i.e. percentage of labeled 
pixels), compared with the 80% accuracy and 26% coverage
from the MISR operational algorithm.
Since two expert labels could differ by around 5%, we have come close
to automating the expert labeling process.

(This talk is based on joint work with Tao Shi at Ohio State University,
Eugene Clothiaux at Penn State University, and Amy Braverman at NASA's 
Jet Propulsion Lab.) 



\end{document}