[Prev][Next][Index]

CSLI Calendar, 21 September 1989, Vol 5:1


       C S L I   C A L E N D A R   O F   P U B L I C   E V E N T S
_____________________________________________________________________________
21 September 1989                   Stanford                    Vol. 5, No. 1
_____________________________________________________________________________

    A weekly publication of the Center for the Study of Language and
Information (CSLI), Ventura Hall, Stanford University, Stanford, CA 94305-4115
			      ____________

	  CSLI ACTIVITIES FOR NEXT THURSDAY, 28 September 1989

12:00 p.m.		TINLunch
      Cordura Hall      Document Image Analysis by Tree Structure Model
      Conference Room   Y. Nishimura, T. Takahashi, and Y. Kobayashi
			ATR Communication Systems Research Laboratories
			(nisimura@atr-sw.atr.co.jp)
			Abstract below

 2:15 p.m.		CSLI Seminar
      Cordura Hall	Models of Rational Agency
      Conference Room	First meeting
			Michael Bratman, Martha Pollack, Stan Rosenschein
			(bratman@csli, pollack@warbucks.ai.sri.com,
			 stan@teleos.com)
			Abstract below

                              ____________

			      ANNOUNCEMENT

This is the first Calendar of the 1989-90 academic year, and we'd like
to welcome everyone back to CSLI's weekly activities, starting next
Thursday.

We would like to thank Emma Pease for having done a great job as
Editor of the Calendar for the last few years.  Emma has taken on new
responsibilities within CSLI, and, starting with this issue, the
Calendar will be edited by Ingrid Deiwiks.

                              ____________
		
			 NEXT WEEK'S TINLUNCH
	   Document Image Analysis by Tree Structure Model
	     Y. Nishimura, T. Takahashi, and Y. Kobayashi
		     (nisimura@atr-sw.atr.co.jp)
			  28 September 1989

In this presentation, I will first outline the organization of ATR
(Advanced Telecommunication Research Institute), which is located in
Kyoto, Japan, and consists of four laboratories:

- ATR Communication Systems Research Laboratories 
- ATR Optical and Radio Communications Research Laboratories
- ATR Interpreting Telephony Research Laboratories
- ATR Auditory and Visual Perception Research Laboratories 

Then, I will introduce the research conducted in our group at ATR
Communication Systems Research Laboratories.  Finally, I will talk
about my research concerning next-generation page readers (OCRs), an
abstract of which is given below.

We propose a document image analysis method that can extract the
logical structure of scanned paper documents to obtain indices such as
titles, author names, etc.  The aim of this research is to develop a
building block for next-generation page readers, which will be able to
capture not only the character codes, but also the layout and logical
structure of documents.

A major problem in document image analysis is segmenting document
images to extract the components (indices).  We implement the
segmentation process as the top-down, model-driven matching process of
the model and the input image.  For this purpose, we introduce a tree
structure model to represent the layout of each document type (such as
title pages of IEEE Trans. papers).  The model also describes elements
of the page such as the body and running heads.  The body is further
divided into the text and other components.

Using the model in a top-down fashion, first the running heads and
running foots are extracted.  Next, the components other than the text
are extracted.  The model describes these components in the way they
are segmented, thus the segmentation result of the input image matches
that of the model.  The text, the part where uniformity is frequently
degraded, is extracted as the remaining part of the page.

We also introduce a model-building process that can be operated by a
novice user.  In an experiment using 115 input documents from 38 types
of scientific paper title pages, every index in 85.2% of the input
documents was correctly extracted.  In a comparison experiment using a
bottom-up segmentation process, the correct extraction rate was only
9.1%.

Applications of the method include document entry without rekeying
for electronic publishing systems, document retrieval, and automatic
indexing of scanned documents for document database systems.

			      ____________

			 NEXT WEEK'S SEMINAR
		      Models of Rational Agency
	  Michael Bratman, Martha Pollack, Stan Rosenschein
			    First meeting
			  28 September 1989

This seminar will survey models of intelligent rational agents in
dynamic, multiagent environments.  We will look at models couched in
terms of commonsense psychological notions like belief, desire,
intention, and plan.  We will explore the relations between such
models and issues of implementation and design as they arise in
artificial intelligence.  In particular, we will be concerned with the
implications of resource limitations for models of rational agency at
various levels of abstraction.